commit 18c602dee4 ("qede: Use NETIF_F_GRO_HW.") introduced
a regression in driver that when xdp program is installed on
qede device, device's aggregation feature (hardware GRO) is not
getting disabled, which is unexpected with xdp.
Fixes: 18c602dee4 ("qede: Use NETIF_F_GRO_HW.")
Signed-off-by: Manish Chopra <manishc@marvell.com>
Signed-off-by: Ariel Elior <aelior@marvell.com>
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Arthur Kiyanovski says:
====================
ena: fixes of interrupt moderation bugs
Differences from V1:
1. Updated default tx interrupt moderation to 64us
2. Added "Fixes:" tags.
3. Removed cosmetic changes that are not relevant for these bug fixes
This patchset includes a couple of fixes of bugs in the implemenation of
interrupt moderation.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Issue 1:
--------
Reproduction steps:
1. sudo ethtool -C eth0 rx-usecs 128
2. sudo ethtool -C eth0 adaptive-rx on
3. sudo ethtool -C eth0 adaptive-rx off
4. ethtool -c eth0
expected output: rx-usecs 128
actual output: rx-usecs 0
Reason for issue:
In stage 3, ethtool userspace calls first the ena_get_coalesce() handler
to get the current value of all properties, and then the ena_set_coalesce()
handler. When ena_get_coalesce() is called the adaptive interrupt
moderation is still on. There is an if in the code that returns the
rx_coalesce_usecs only if the adaptive interrupt moderation is off.
And since it is still on, rx_coalesce_usecs is not set, meaning it
stays 0.
Solution to issue:
Remove this if static interrupt moderation intervals have nothing to do
with dynamic ones.
Issue 2:
--------
Reproduction steps:
1. sudo ethtool -C eth0 adaptive-rx on
2. sudo ethtool -C eth0 rx-usecs 128
3. ethtool -c eth0
expected output: rx-usecs 128
actual output: rx-usecs 0
Reason for issue:
In stage 2, when ena_set_coalesce() is called, the handler tests if
rx adaptive interrupt moderation is on, and if it is, it returns before
getting to the part in the function that sets the rx non-adaptive
interrupt moderation interval.
Solution to issue:
Remove the return from the function when rx adaptive interrupt moderation
is on.
Also cleaned up the fixed code in ena_set_coalesce by grouping together
adaptive interrupt moderation toggling, and using && instead of nested
ifs.
Fixes: b3db86dc4b ("net: ena: reimplement set/get_coalesce()")
Fixes: 0eda847953 ("net: ena: fix retrieval of nonadaptive interrupt moderation intervals")
Fixes: 1738cd3ed3 ("net: ena: Add a driver for Amazon Elastic Network Adapters (ENA)")
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Current default non-adaptive tx interrupt moderation interval is 196 us.
This value is too high and might cause the tx queue to fill up.
In this commit we set the default non-adaptive tx interrupt moderation
interval to 64 us in order to:
1. Reduce the probability of the queue filling-up (when compared to the
current default value of 196 us).
2. Reduce unnecessary tx interrupt overhead (which happens if we set the
default tx interval to 0).
We determined experimentally that 64 us is an optimal value that
reduces interrupt rate by more than 20% without affecting performance.
Fixes: 1738cd3ed3 ("net: ena: Add a driver for Amazon Elastic Network Adapters (ENA)")
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the reboot_event handler, unregister the ib devices and enable
the IB layer to release the devices before the reboot.
Fixes: a33a803cfe ("net/smc: guarantee removal of link groups in reboot")
Signed-off-by: Karsten Graul <kgraul@linux.ibm.com>
Reviewed-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The current implementation of "stmmac_dt_phy" function initializes
the MDIO platform bus data, even in the absence of PHY. This fix
will skip MDIO initialization if there is no PHY present.
Fixes: 7437127 ("net: stmmac: Convert to phylink and remove phylib logic")
Acked-by: Jayati Sahu <jayati.sahu@samsung.com>
Signed-off-by: Sriram Dash <sriram.dash@samsung.com>
Signed-off-by: Padmanabhan Rajanbabu <p.rajanbabu@samsung.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Huazhong Tan says:
====================
net: hns3: misc updates for -net-next
This series includes some misc updates for the HNS3 ethernet driver.
[patch 1] adds FE bit check before calling hns3_add_frag().
[patch 2] removes an unnecessary lock.
[patch 3] adds a little optimization for CMDQ uninitialization.
[patch 4] refactors the dump of FD tcams.
[patch 5] implements ndo_features_check ops.
[patch 6] adds some VF VLAN information for command "ip link show".
[patch 7] adds a debug print.
[patch 8] modifies the timing of print misc interrupt status when
handling hardware error event.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Printing misc interrupt status of hardware error event in the
IRQ handler is unnecessary, since hclge_handle_hw_msix_error()
will print out the detail information for this hardware error
when handling success. So, this patch removes the print in
IRQ handler, and prints it when hclge_handle_hw_msix_error()
fails.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since the mapping can be overwritten, when fail to get
the chain between vector and ring, we should go on to
deal with the remaining options. For debugging, this
patch adds log info for this failure.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds some VF VLAN information for command "ip link show".
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The function netif_skb_features() will disable the TSO feature
by using dflt_features_check() if the driver does not implement
ndo_features_check ops, which may cause performance degradation
problem when hns3 hardware can do multiple tagged TSO.
Also, the HNS3 hardware only supports checksum on the SKB with
a max header len of 480 bytes, so remove the checksum and TSO
related features when the header len is over 480 bytes.
Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, the dump FD tcam mode in debugfs is to query all FD tcams,
including empty rules, which is unnecessary. This patch modify to
find the position of useful rules before dump FD tcam, so that it does
not need to query empty rules.
This patch also modifies some help information in debugfs.
Signed-off-by: Yufeng Mo <moyufeng@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When uninitializing CMDQ, HCLGE_STATE_CMD_DISABLE will
be set up firstly, then the driver does not send command
anymore. So, hclge_free_cmd_desc can be called without
holding ring->lock. hclge_destroy_cmd_queue() and
hclge_destroy_queue() are unnecessary now, so removes them,
the VF driver has implemented currently.
BTW, the VF driver should set up HCLGEVF_STATE_CMD_DISABLE
as well in the hclgevf_cmd_uninit(), just likes what the PF
driver does.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Mutex vport_cfg_mutex has been used to protect uc_mac_list,
mc_mac_list and vlan_list from being modified by unloading
or reset task at the same time. But now unloading will
set up HCLGE_STATE_REMOVING flag and call cancel_work_sync to
break down this race condition, so this mutex is unnecessary.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A BD with FE bit means that it is the last BD of a packet,
currently the FE bit is checked before calling hns3_add_frag(),
which is unnecessary because the FE bit may have been checked
in some case.
This patch checks the FE bit before calling hns3_add_frag()
after processing the first BD of a SKB and adjust the location
of memcpy() to reduce duplication.
Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a frame with NULL DSAP is received, llc_station_rcv is called.
In turn, llc_stat_ev_rx_null_dsap_xid_c is called to check if it is a NULL
XID frame. The return statement of llc_stat_ev_rx_null_dsap_xid_c returns 1
when the incoming frame is not a NULL XID frame and 0 otherwise. Hence, a
NULL XID response is returned unexpectedly, e.g. when the incoming frame is
a NULL TEST command.
To fix the error, simply remove the conditional operator.
A similar error in llc_stat_ev_rx_null_dsap_test_c is also fixed.
Signed-off-by: Chan Shu Tak, Alex <alexchan@task.com.hk>
Signed-off-by: David S. Miller <davem@davemloft.net>
To enable iproute2/tipc to generate backwards compatible
printouts and validate command parameters for nodes using a
<z.c.n> node address, it needs to be able to read the legacy
address flag from the kernel. The legacy address flag records
the way in which the node identity was originally specified.
The legacy address flag is requested by the netlink message
TIPC_NL_ADDR_LEGACY_GET. If the flag is set the attribute
TIPC_NLA_NET_ADDR_LEGACY is set in the return message.
Signed-off-by: John Rutherford <john.rutherford@dektech.com.au>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When doing stress test, we get the following trace:
kernel BUG at lib/dynamic_queue_limits.c:26!
Internal error: Oops - BUG: 0 [#1] SMP ARM
Modules linked in: hip04_eth
CPU: 0 PID: 2003 Comm: tDblStackPcap0 Tainted: G O L 4.4.197 #1
Hardware name: Hisilicon A15
task: c3637668 task.stack: de3bc000
PC is at dql_completed+0x18/0x154
LR is at hip04_tx_reclaim+0x110/0x174 [hip04_eth]
pc : [<c041abfc>] lr : [<bf0003a8>] psr: 800f0313
sp : de3bdc2c ip : 00000000 fp : c020fb10
r10: 00000000 r9 : c39b4224 r8 : 00000001
r7 : 00000046 r6 : c39b4000 r5 : 0078f392 r4 : 0078f392
r3 : 00000047 r2 : 00000000 r1 : 00000046 r0 : df5d5c80
Flags: Nzcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user
Control: 32c5387d Table: 1e189b80 DAC: 55555555
Process tDblStackPcap0 (pid: 2003, stack limit = 0xde3bc190)
Stack: (0xde3bdc2c to 0xde3be000)
[<c041abfc>] (dql_completed) from [<bf0003a8>] (hip04_tx_reclaim+0x110/0x174 [hip04_eth])
[<bf0003a8>] (hip04_tx_reclaim [hip04_eth]) from [<bf0012c0>] (hip04_rx_poll+0x20/0x388 [hip04_eth])
[<bf0012c0>] (hip04_rx_poll [hip04_eth]) from [<c04c8d9c>] (net_rx_action+0x120/0x374)
[<c04c8d9c>] (net_rx_action) from [<c021eaf4>] (__do_softirq+0x218/0x318)
[<c021eaf4>] (__do_softirq) from [<c021eea0>] (irq_exit+0x88/0xac)
[<c021eea0>] (irq_exit) from [<c0240130>] (msa_irq_exit+0x11c/0x1d4)
[<c0240130>] (msa_irq_exit) from [<c0267ba8>] (__handle_domain_irq+0x110/0x148)
[<c0267ba8>] (__handle_domain_irq) from [<c0201588>] (gic_handle_irq+0xd4/0x118)
[<c0201588>] (gic_handle_irq) from [<c0558360>] (__irq_svc+0x40/0x58)
Exception stack(0xde3bdde0 to 0xde3bde28)
dde0: 00000000 00008001 c3637668 00000000 00000000 a00f0213 dd3627a0 c0af6380
de00: c086d380 a00f0213 c0a22a50 de3bde6c 00000002 de3bde30 c0558138 c055813c
de20: 600f0213 ffffffff
[<c0558360>] (__irq_svc) from [<c055813c>] (_raw_spin_unlock_irqrestore+0x44/0x54)
Kernel panic - not syncing: Fatal exception in interrupt
Pre-modification code:
int hip04_mac_start_xmit(struct sk_buff *skb, struct net_device *ndev)
{
[...]
[1] priv->tx_head = TX_NEXT(tx_head);
[2] count++;
[3] netdev_sent_queue(ndev, skb->len);
[...]
}
An rx interrupt occurs if hip04_mac_start_xmit just executes to the line 2,
tx_head has been updated, but corresponding 'skb->len' has not been
added to dql_queue.
And then
hip04_mac_interrupt->__napi_schedule->hip04_rx_poll->hip04_tx_reclaim
In hip04_tx_reclaim, because tx_head has been updated,
bytes_compl will plus an additional "skb-> len"
which has not been added to dql_queue. And then
trigger the BUG_ON(bytes_compl > num_queued - dql->num_completed).
To solve the problem described above, we put
"netdev_sent_queue(ndev, skb->len);"
before
"priv->tx_head = TX_NEXT(tx_head);"
Fixes: a41ea46a9a ("net: hisilicon: new hip04 ethernet driver")
Signed-off-by: Jiangfeng Xiao <xiaojiangfeng@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Garzarella says:
====================
VSOCK: add vsock_test test suite
The vsock_diag.ko module already has a test suite but the core AF_VSOCK
functionality has no tests. This patch series adds several test cases that
exercise AF_VSOCK SOCK_STREAM socket semantics (send/recv, connect/accept,
half-closed connections, simultaneous connections).
The v1 of this series was originally sent by Stefan.
v3:
- Patch 6:
* check the byte received in the recv_byte()
* use send(2)/recv(2) instead of write(2)/read(2) to test also flags
(e.g. MSG_PEEK)
- Patch 8:
* removed unnecessary control_expectln("CLOSED") [Stefan].
- removed patches 9,10,11 added in the v2
- new Patch 9 add parameters to list and skip tests (e.g. useful for vmci
that doesn't support half-closed socket in the host)
- new Patch 10 prints a list of options in the help
- new Patch 11 tests MSG_PEEK flags of recv(2)
v2: https://patchwork.ozlabs.org/cover/1140538/
v1: https://patchwork.ozlabs.org/cover/847998/
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Test if the MSG_PEEK flags of recv(2) works as expected.
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since we now have several options, in the help we print the list
of all supported options and a brief description of them.
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some tests can fail with transports that have a slightly
different behavior, so let's add the possibility to specify
which tests to skip.
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Before check if a send returns -EPIPE, we need to make sure the
connection is closed.
To do that, we use epoll API to wait EPOLLRDHUP or EPOLLHUP events
on the socket.
Reported-by: Jorgen Hansen <jhansen@vmware.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The vsock_test.c program runs a test suite of AF_VSOCK test cases.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Test cases will want to transfer data. This patch adds utility
functions to do this.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
See code comment for details.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Many test cases will need to connect to the server or accept incoming
connections. This patch extracts these operations into utility
functions that can be reused.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move useful functions into a separate file in preparation for more
vsock test programs.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The vsock_diag_test program directly included ../../../include/uapi/
headers from the source tree. Tests are supposed to use the
usr/include/linux/ headers that have been prepared with make
headers_install instead.
Suggested-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove special taglen define KSZ8795_INGRESS_TAG_LEN
and use generic KSZ_INGRESS_TAG_LEN instead.
Signed-off-by: Michael Grzeschik <m.grzeschik@pengutronix.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Julian Wiedmann says:
====================
s390/qeth: fixes 2019-12-18
please apply the following patch series to your net tree.
This brings two fixes for initialization / teardown issues, and one
ENOTSUPP cleanup.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
ENOTSUPP is not uapi, use EOPNOTSUPP instead.
Fixes: d66cb37e96 ("qeth: Add new priority queueing options")
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When managing the promiscuous mode during an RX modeset, qeth caches the
current HW state to avoid repeated programming of the same state on each
modeset.
But while tearing down a device, we forget to clear the cached state. So
when the device is later set online again, the initial RX modeset
doesn't program the promiscuous mode since we believe it is already
enabled.
Fix this by clearing the cached state in the tear-down path.
Note that for the SBP variant of promiscuous mode, this accidentally
works right now because we unconditionally restore the SBP role while
re-initializing.
Fixes: 4a71df5004 ("qeth: new qeth device driver")
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Reviewed-by: Alexandra Winter <wintera@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Along with z/VM NICs, there's additional device types that only support
a specific transport mode (eg. external-bridged IQD).
Identify the corresponding error code, and raise a fitting error message
so that the user knows to adjust their device configuration.
On top of that also fix the subsequent error path, so that the rejected
cmd doesn't need to wait for a timeout but gets cancelled straight away.
Fixes: 4a71df5004 ("qeth: new qeth device driver")
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Oleksij Rempel says:
====================
add dsa switch support for ar9331
changes v6:
- remove ag71xx changes from this patch set. It needs more work.
- ar9331: fix register definition and add ASCII art switch documentation.
changes v6:
- rebase against net-next
changes v5:
- remote support for port5. The effort of using this port is
questionable. Currently, it is better to not use it at all, then
adding buggy support.
- remove port enable call back. There is nothing what we actually need
to enable.
- rebase it against v5.5-rc1
changes v4:
- ag71xx: ag71xx_mac_validate fix always false comparison (&& -> ||)
- tag_ar9331: use skb_pull_rcsum() instead of skb_pull().
- tag_ar9331: drop skb_set_mac_header()
changes v3:
- ag71xx: ag71xx_mac_config: ignore MLO_AN_INBAND mode. It is not
supported by HW and SW.
- ag71xx: ag71xx_mac_validate: return all supported bits on
PHY_INTERFACE_MODE_NA
changes v2:
- move Atheros AR9331 TAG format to separate patch
- use netdev_warn_once in the tag driver to reduce potential message spam
- typo fixes
- reorder tag driver alphabetically
- configure switch to maximal frame size
- use mdiobus_read/write
- fail if mdio sub node is not found
- add comment for post reset state
- remove deprecated comment about device id
- remove phy-handle option for node with fixed-link
- ag71xx: set 1G support only for GMII mode
This patch series provides dsa switch support for Atheros ar9331 WiSoC.
As side effect ag71xx needed to be ported to phylink to make the switch
driver (as well phylink based) work properly.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Provide basic support for Atheros AR9331 built-in switch. So far it
works as port multiplexer without any hardware offloading support.
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Vivien Didelot <vivien.didelot@gmail.com>
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add support for tag format used in Atheros AR9331 built-in switch.
Reviewed-by: Vivien Didelot <vivien.didelot@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add switch node supported by dsa ar9331 driver.
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Atheros AR9331 has built-in 5 port switch. The switch can be configured
to use all 5 or 4 ports. One of built-in PHYs can be used by first built-in
ethernet controller or to be used directly by the switch over second ethernet
controller.
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
When sample xdp_redirect_cpu was converted to use libbpf, the
tracepoints used by this sample were not getting attached automatically
like with bpf_load.c. The BPF-maps was still getting loaded, thus
nobody notice that the tracepoints were not updating these maps.
This fix doesn't use the new skeleton code, as this bug was introduced
in v5.1 and stable might want to backport this. E.g. Red Hat QA uses
this sample as part of their testing.
Fixes: bbaf6029c4 ("samples/bpf: Convert XDP samples to libbpf usage")
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/157685877642.26195.2798780195186786841.stgit@firesoul
Jay Jayatheerthan says:
====================
This series of patches enhances xdpsock application with command line
parameters to set transmit packet size and fill pattern among other options.
The application has also been enhanced to use Linux Ethernet/IP/UDP header
structs and calculate IP and UDP checksums.
I have measured the performance of the xdpsock application before and after
this patch set and have not been able to detect any difference.
Packet Size:
------------
There is a new option '-s' or '--tx-pkt-size' to specify the transmit packet
size. It ranges from 47 to 4096 bytes. Default packet size is 64 bytes
which is same as before.
Fill Pattern:
-------------
The transmit UDP payload fill pattern is specified using '-P' or
'--tx-pkt-pattern'option. It is an unsigned 32 bit field and defaulted
to 0x12345678.
Packet Count:
-------------
The number of packets to send is specified using '-C' or '--tx-pkt-count'
option. If it is not specified, the application sends packets forever.
Batch Size:
-----------
The batch size for transmit, receive and l2fwd features of the application is
specified using '-b' or '--batch-size' options. Default value when this option
is not provided is 64 (same as before).
Duration:
---------
The application supports '-d' or '--duration' option to specify number of
seconds to run. This is used in tx, rx and l2fwd features. If this option is
not provided, the application runs for ever.
====================
Tested-by: Björn Töpel <bjorn.topel@intel.com>
Acked-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The UDP payload fill pattern can be specified using '-P' or '--tx-pkt-pattern'
option. It is an unsigned 32 bit field and defaulted to 0x12345678.
The IP and UDP checksum is calculated by the code as per the content of
the packet before transmission.
Signed-off-by: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191220085530.4980-7-jay.jayatheerthan@intel.com
New option '-s' or '--tx-pkt-size' has been added to specify the transmit
packet size. The packet size ranges from 47 to 4096 bytes. When this
option is not provided, it defaults to 64 byte packet.
The code uses struct ethhdr, struct iphdr and struct udphdr to form the
transmit packet. The MAC address, IP address and UDP ports are set to default
values.
The code calculates IP and UDP checksums before sending the packet.
Checksum calculation code in Linux kernel is used for this purpose.
The Ethernet FCS is not filled by the code.
Signed-off-by: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191220085530.4980-6-jay.jayatheerthan@intel.com
Use '-C' or '--tx-pkt-count' to specify number of packets to send.
If it is not specified, the application sends packets forever. If packet
count is not a multiple of batch size, last batch sent is less than the
batch size.
Signed-off-by: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191220085530.4980-5-jay.jayatheerthan@intel.com
New option to specify batch size for tx, rx and l2fwd has been added. This
allows fine tuning to maximize performance. It is specified using '-b' or
'--batch_size' options. When not specified default is 64.
Signed-off-by: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191220085530.4980-4-jay.jayatheerthan@intel.com
Add code to do cleanup for signals and application completion in a unified
fashion. The signal handler sets benckmark_done flag terminating the
threads. The cleanup is called before returning from main() function.
Signed-off-by: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191220085530.4980-3-jay.jayatheerthan@intel.com
The application now supports '-d' or '--duration' option to specify number of
seconds to run. This is used in tx, rx and l2fwd features. If this option is
not provided, the application runs forever.
Signed-off-by: Jay Jayatheerthan <jay.jayatheerthan@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191220085530.4980-2-jay.jayatheerthan@intel.com
It's follow-up for discussion [1]
CHECK and CHECK_FAIL macros in test_progs.h can affect errno in some
circumstances, e.g. if some code accidentally closes stdout. It makes
checking errno in patterns like this unreliable:
if (CHECK(!bpf_prog_attach_xattr(...), "tag", "msg"))
goto err;
CHECK_FAIL(errno != ENOENT);
, since by CHECK_FAIL time errno could be affected not only by
bpf_prog_attach_xattr but by CHECK as well.
Fix it by saving and restoring errno in the macros. There is no "Fixes"
tag since no problems were discovered yet and it's rather precaution.
test_progs was run with this change and no difference was identified.
[1] https://lore.kernel.org/bpf/20191219210907.GD16266@rdna-mbp.dhcp.thefacebook.com/
Signed-off-by: Andrey Ignatov <rdna@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20191220000511.1684853-1-rdna@fb.com
Magnus Karlsson says:
====================
This patch set cleans up the ring access functions of AF_XDP in hope
that it will now be easier to understand and maintain. I used to get a
headache every time I looked at this code in order to really understand it,
but now I do think it is a lot less painful.
The code has been simplified a lot and as a bonus we get better
performance in nearly all cases. On my new 2.1 GHz Cascade Lake
machine with a standard default config plus AF_XDP support and
CONFIG_PREEMPT on I get the following results in percent performance
increases with this patch set compared to without it:
Zero-copy (-N):
rxdrop txpush l2fwd
1 core: -2% 0% 3%
2 cores: 4% 0% 3%
Zero-copy with poll() (-N -p):
rxdrop txpush l2fwd
1 core: 3% 0% 1%
2 cores: 21% 0% 9%
Skb mode (-S):
Shows a 0% to 5% performance improvement over the same benchmarks as
above.
Here 1 core means that we are running the driver processing and the
application on the same core, while 2 cores means that they execute on
separate cores. The applications are from the xdpsock sample app.
On my older 2.0 Ghz Broadwell machine that I used for the v1, I get
the following results:
Zero-copy (-N):
rxdrop txpush l2fwd
1 core: 4% 5% 4%
2 cores: 1% 0% 2%
Zero-copy with poll() (-N -p):
rxdrop txpush l2fwd
1 core: 1% 3% 3%
2 cores: 22% 0% 5%
Skb mode (-S):
Shows a 0% to 1% performance improvement over the same benchmarks as
above.
When a results says 21 or 22% better, as in the case of poll mode with
2 cores and rxdrop, my first reaction is that it must be a
bug. Everything else shows between 0% and 5% performance
improvement. What is giving rise to 22%? A quick bisect indicates that
it is patches 2, 3, 4, 5, and 6 that are giving rise to most of this
improvement. So not one patch in particular, but something around 4%
improvement from each one of them. Note that exactly this benchmark
has previously had an extraordinary slow down compared to when running
without poll syscalls. For all the other poll tests above, the
slowdown has always been around 4% for using poll syscalls. But with
the bad performing test in question, it was above 25%. Interestingly,
after this clean up, the slow down is 4%, just like all the other poll
tests. Please take an extra peek at this so I have not messed up
something.
The 0% for several txpush results are due to the test bottlenecking on
a non-CPU HW resource. If I eliminated that bottleneck on my system, I
would expect to see an increase there too.
Changes v1 -> v2:
* Corrected textual errors in the commit logs (Sergei and Martin)
* Fixed the functions that detect empty and full rings so that they
now operate on the global ring state (Maxim)
This patch has been applied against commit a352a82496 ("Merge branch 'libbpf-extern-followups'")
Structure of the patch set:
Patch 1: Eliminate the lazy update threshold used when preallocating
entries in the completion ring
Patch 2: Simplify the detection of empty and full rings
Patch 3: Consolidate the two local producer pointers into one
Patch 4: Standardize the naming of the producer ring access functions
Patch 5: Eliminate the Rx batch size used for the fill ring
Patch 6: Simplify the functions xskq_nb_avail and xskq_nb_free
Patch 7: Simplify and standardize the naming of the consumer ring
access functions
Patch 8: Change the names of the validation functions to improve
readability and also the return value of these functions
Patch 9: Change the name of xsk_umem_discard_addr() to
xsk_umem_release_addr() to better reflect the new
names. Requires a name change in the drivers that support AF_XDP
zero-copy.
Patch 10: Remove unnecessary READ_ONCE of data in the ring
Patch 11: Add overall function naming comment and reorder the functions
for easier reference
Patch 12: Use the struct_size helper function when allocating rings
====================
Reviewed-by: Björn Töpel <bjorn.topel@intel.com>
Tested-by: Björn Töpel <bjorn.topel@intel.com>
Acked-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
