Michal Krawczyk [Wed, 8 Apr 2020 08:29:08 +0000 (10:29 +0200)]
net/ena: set IO ring size to valid value
IO rings were configured with the maximum allowed size for the Tx/Rx
rings. However, the application could decide to create smaller rings.
This patch is using value stored in the ring instead of the value from
the adapter which is indicating the maximum allowed value.
Fixes: df238f84c0a2 ("net/ena: recreate HW IO rings on start and stop") Cc: stable@dpdk.org Signed-off-by: Michal Krawczyk <mk@semihalf.com> Reviewed-by: Igor Chauskin <igorch@amazon.com> Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Michal Krawczyk [Wed, 8 Apr 2020 08:29:06 +0000 (10:29 +0200)]
net/ena/base: fix indentation of multiple defines
As the alignment of the defines wasn't valid, it was removed at all, so
instead of using multiple spaces or tabs, the single space after define
name is being used.
Fixes: 99ecfbf845b3 ("ena: import communication layer") Signed-off-by: Michal Krawczyk <mk@semihalf.com> Reviewed-by: Igor Chauskin <igorch@amazon.com> Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Michal Krawczyk [Wed, 8 Apr 2020 08:29:05 +0000 (10:29 +0200)]
net/ena/base: fix types for printing timestamps
Because ena_com is being used by multiple platforms which are using
different C versions, PRIu64 cannot be used directly and must be defined
in the platform file.
Fixes: b2b02edeb0d6 ("net/ena/base: upgrade HAL for new HW features") Signed-off-by: Michal Krawczyk <mk@semihalf.com> Reviewed-by: Igor Chauskin <igorch@amazon.com> Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Michal Krawczyk [Wed, 8 Apr 2020 08:29:04 +0000 (10:29 +0200)]
net/ena/base: use 48-bit memory addresses
ENA device is using 48-bit memory for IO. Because of that, the upper
limit had to be updated.
From the driver perspective, it's just a cosmetic change to make
definition of the structure 'ena_common_mem_addr' more descriptive and
the address value was verified anyway for the valid range in the
function 'ena_com_mem_addr_set()'.
Signed-off-by: Michal Krawczyk <mk@semihalf.com> Reviewed-by: Igor Chauskin <igorch@amazon.com> Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Michal Krawczyk [Wed, 8 Apr 2020 08:29:02 +0000 (10:29 +0200)]
net/ena/base: fix indentation in CQ polling
The spaces instead of tabs were used for the indent.
Fixes: 3adcba9a8987 ("net/ena: update HAL to the newer version") Signed-off-by: Michal Krawczyk <mk@semihalf.com> Reviewed-by: Igor Chauskin <igorch@amazon.com> Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Michal Krawczyk [Wed, 8 Apr 2020 08:29:01 +0000 (10:29 +0200)]
net/ena/base: fix documentation of functions
The documentation format was aligned and few typos were fixed.
Fixes: 99ecfbf845b3 ("ena: import communication layer") Signed-off-by: Michal Krawczyk <mk@semihalf.com> Reviewed-by: Igor Chauskin <igorch@amazon.com> Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Michal Krawczyk [Wed, 8 Apr 2020 08:29:00 +0000 (10:29 +0200)]
net/ena/base: add accelerated LLQ mode
In order to use the accelerated LLQ (Low-lateny queue) mode, the driver
must limit the Tx burst and be aware that the device has the meta
caching disabled. In that situation, the meta descriptor must be valid
on each Tx packet.
Signed-off-by: Michal Krawczyk <mk@semihalf.com> Reviewed-by: Igor Chauskin <igorch@amazon.com> Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Michal Krawczyk [Wed, 8 Apr 2020 08:28:56 +0000 (10:28 +0200)]
net/ena/base: fix testing for supported hash function
There was a bug in ena_com_fill_hash_function(), which was causing bit to
be shifted left one bit too much.
To fix that, the ENA_FFS macro is being used (returning the location of
the first bit set), hash_function value is being subtracted by 1 if any
hash function is supported by the device and BIT macro is used for
shifting for better verbosity.
Fixes: 99ecfbf845b3 ("ena: import communication layer") Cc: stable@dpdk.org Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Igor Chauskin [Wed, 8 Apr 2020 08:28:54 +0000 (10:28 +0200)]
net/ena/base: prevent allocation of zero sized memory
rte_memzone_reserve() will reserve the biggest contiguous memzone
available if received 0 as size param.
Fixes: 9ba7981ec992 ("ena: add communication layer for DPDK") Cc: stable@dpdk.org Signed-off-by: Igor Chauskin <igorch@amazon.com> Reviewed-by: Michal Krawczyk <mk@semihalf.com> Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Igor Chauskin [Wed, 8 Apr 2020 08:28:53 +0000 (10:28 +0200)]
net/ena/base: make allocation macros thread-safe
Memory allocation region id could possibly be non-unique
due to non-atomic increment, causing allocation failure.
Fixes: 9ba7981ec992 ("ena: add communication layer for DPDK") Cc: stable@dpdk.org Signed-off-by: Igor Chauskin <igorch@amazon.com> Reviewed-by: Michal Krawczyk <mk@semihalf.com> Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Michal Krawczyk [Wed, 8 Apr 2020 08:28:52 +0000 (10:28 +0200)]
net/ena: ensure Rx buffer size is at least 1400B
Some of the ENA devices can't handle buffers which are smaller than a
1400B. Because of this limitation, size of the buffer is being checked
and limited during the Rx queue setup.
If it's below the allowed value, PMD won't finish it's configuration
successfully..
Signed-off-by: Michal Krawczyk <mk@semihalf.com> Reviewed-by: Igor Chauskin <igorch@amazon.com> Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Update a switch rule' action from "to VSI" to "to VSI List"
should only happen when the same rule has been programmed with
a different fwd destination. This is already handled by below
code block:
m_entry = ice_find_adv_rule_entry(...)
if (m_entry) {
...
ice_adv_add_update_vsi_list(...)
}
The following ice_update_pkt_fwd_rule is unnecessary and should be
removed due to:
1) If a switch rule's action is still to VSI, which means, it is
the first time be issued, we don't need to update it "to VSI
List."
2) Actually the implementation does not match the comment, it still
update the rule with "to VSI" action.
Fixes: fed0c5ca5f19 ("net/ice/base: support programming a new switch recipe") Cc: stable@dpdk.org Signed-off-by: Qi Zhang <qi.z.zhang@intel.com> Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
Lunyuan Cui [Thu, 2 Apr 2020 07:58:47 +0000 (07:58 +0000)]
net/i40e: enable MAC address as flow director input set
Enable source MAC address and destination MAC address as FDIR's
input set for ipv4-other, ipv4-udp and ipv4-tcp. When OVS-DPDK is
working as a pure L2 switch, enable MAC address as FDIR input set
with Mark+RSS action would help the performance speed up. And FVL
FDIR supports to change input set with MAC address.
Signed-off-by: Lunyuan Cui <lunyuanx.cui@intel.com> Acked-by: Beilei Xing <beilei.xing@intel.com>
Yunjian Wang [Tue, 7 Apr 2020 11:37:27 +0000 (19:37 +0800)]
net/nfp: fix dangling pointer on probe failure
When nfp_pf_create_dev() is cleaning up, it does not correctly set
the dev_private variable to NULL, which will lead to a double free.
Fixes: ef28aa96e53b ("net/nfp: support multiprocess") Cc: stable@dpdk.org Signed-off-by: Yunjian Wang <wangyunjian@huawei.com> Acked-by: Heinrich Kuhn <heinrich.kuhn@netronome.com>
Add the previous prototype for the 'profile_hook_rx_burst_cb' function
to fix the compiler warning when the DPDK is built with
'RTE_ETHDEV_PROFILE_WITH_VTUNE' config option enabled:
/home/dpdk/lib/librte_ethdev/ethdev_profile.c:17:1: warning:
no previous prototype for profile_hook_rx_burst_cb [-Wmissing-prototypes]
Since the ring buffer with host is shared for both transmit
completions and receive packets, it is possible that transmitter
could get starved if receive ring gets full.
Better to process all outstanding events which frees up transmit
buffer slots, even if means dropping some packets.
Fixes: 7e6c82430702 ("net/netvsc: avoid over filling Rx descriptor ring") Cc: stable@dpdk.org Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
The netvsc PMD was putting the mac address in private data but the
core rte_ethdev doesn't allow that it. It has to be in rte_malloc'd
memory or a message will be printed on shutdown/close.
EAL: Invalid memory
Fixes: f8279f47dd89 ("net/netvsc: fix crash in secondary process") Cc: stable@dpdk.org Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
net/netvsc: split send buffers from Tx descriptors
The VMBus has reserved transmit area (per device) and transmit
descriptors (per queue). The previous code was always having a 1:1
mapping between send buffers and descriptors.
This can lead to one queue starving another and also buffer bloat.
Change to working more like FreeBSD where there is a pool of transmit
descriptors per queue. If send buffer is not available then no
aggregation happens but the queue can still drain.
net/netvsc: handle Rx packets during multi-channel setup
It is possible for a packet to arrive during the configuration
process when setting up multiple queue mode. This would cause
configure to fail; fix by just ignoring receive packets while
waiting for control commands.
Use the receive ring lock to avoid possible races between
oddly behaved applications doing rx_burst and control operations
concurrently.
Dekel Peled [Wed, 25 Mar 2020 08:12:31 +0000 (10:12 +0200)]
app/testpmd: enhance GTP support
This patch adds CLI option to enter the v_pt_rsv_flags value for GTP
flow pattern item.
It also adds GTP as valid item in raw_encap and raw_decap setting.
Signed-off-by: Dekel Peled <dekelp@mellanox.com> Acked-by: Ori Kam <orika@mellanox.com>
After VF reset, VF's VSI number may be changed,
the switch rule which forwards packet to the old
VSI number should be redirected to the new VSI
number.
The input set for inner type of vlan item should
be ICE_INSET_ETHERTYPE, not ICE_INSET_VLAN_OUTER.
This mac vlan filter is also part of DCF switch filter.
This patch add switch filter support for PFCP packets,
it enable switch filter to direct IPv4/IPv6 packets with
PFCP session or node payload to specific action.
net/ice: change switch parser to support flexible mask
DCF need to make configuration of flexible mask, that is to say
some input set mask may be not 0xFFFF type all one bit. In order
to direct L2/IP multicast packets, the mask for source IP maybe
0xF0000000, this patch enable switch filter parser for it.
DCF on CVL is a control plane VF which take the responsibility to
configure all the PF/global resources, this patch add support DCF
on to program forward rule to direct packets to VFs.
When an application invokes rte_eth_dev_configure consecutively without
setting up Rx/Tx queues, it will incorrectly return error while trying
to restore Rx/Tx queue configuration.
Fix configuration sequence by checking if any Rx/Tx queues are
previously configured before trying to restore them.
Krzysztof Kanas [Sun, 5 Apr 2020 21:16:57 +0000 (02:46 +0530)]
net/octeontx2: add TM capability
Add Traffic Management capability callbacks to provide
global, level and node capabilities. This patch also
adds documentation on Traffic Management Support.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com> Signed-off-by: Krzysztof Kanas <kkanas@marvell.com> Acked-by: Jerin Jacob <jerinj@marvell.com>
Krzysztof Kanas [Fri, 3 Apr 2020 08:52:15 +0000 (14:22 +0530)]
net/octeontx2: add Tx queue rate limit
Add Tx queue ratelimiting support. This support is mutually
exclusive with TM support i.e when TM is configured, tx queue
ratelimiting config is no more valid.
Signed-off-by: Krzysztof Kanas <kkanas@marvell.com> Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com> Acked-by: Jerin Jacob <jerinj@marvell.com>
Add dynamic parent and shaper update callbacks that
can be used to change RR Quantum or PIR/CIR rate dynamically
post hierarchy commit. Dynamic parent update callback only
supports updating RR quantum of a given child with respect to
its parent. There is no support yet to change priority or parent
itself.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com> Signed-off-by: Krzysztof Kanas <kkanas@marvell.com>
Adds support to Traffic Management callbacks "node_add"
and "node_delete". These callbacks doesn't support
dynamic node addition or deletion post hierarchy commit.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com> Signed-off-by: Krzysztof Kanas <kkanas@marvell.com>
Modify resource allocation and freeing logic to support
dynamic topology commit while to traffic is flowing.
This patch also modifies SQ flush to timeout based on minimum shaper
rate configured. SQ flush is further split to pre/post
functions to adhere to HW spec of 96XX C0.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com> Signed-off-by: Krzysztof Kanas <kkanas@marvell.com>
net/octeontx2: enable error and RAS interrupt in configure
Patch adds routines to set/clear nix lf error & ras interrupt enable
registers. These nix lf error interrupts get triggered if there are
any failures during nix lf configuration. This interrupts are enabled
before any hardware configurations initiated on the allocated nix lf.
Signed-off-by: Vamsi Attunuru <vattunuru@marvell.com> Acked-by: Andrzej Ostruszka <aostruszka@marvell.com> Acked-by: Jerin Jacob <jerinj@marvell.com>
Harman Kalra [Mon, 16 Mar 2020 09:33:44 +0000 (15:03 +0530)]
net/octeontx: support Rx/Tx checksum offload
This patch implements rx/tx checksum offload. In case of
wrong checksum received (inner/outer l3/l4) it reports the
corresponding layer which has bad checksum and also corrects
it if hw checksum is enabled on tx side.
Harman Kalra [Mon, 16 Mar 2020 09:33:42 +0000 (15:03 +0530)]
net/octeontx: support set link up/down
Adding support for setting link up/down eth operation.
It is used to enable disable lmac. Also implemented a
poll function for getting the link status at regular
intervals.
Vamsi Attunuru [Mon, 16 Mar 2020 09:33:41 +0000 (15:03 +0530)]
net/octeontx: support VLAN filter offload
Patch adds support for vlan filter offload support.
MBOX messages for vlan filter on/off and vlan filter
entry add/rm are added to configure PCAM entries to
filter out the vlan traffic on a given port.
Patch also defines rx_offload_flag for vlan filtering.
Kiran Kumar K [Sat, 7 Mar 2020 09:56:53 +0000 (15:26 +0530)]
net/octeontx2: offload bad L2/L3/L4 UDP lengths detection
Octeontx2 HW has support for detecting the bad L2/L3/L4 UDP lengths.
Since DPDK does not have specific error flag for this, exposing it
as bad checksum failure in mbuff:ol_flags to leverage this feature.
These errors will be propagated to the ol_flags as follows.
Amit Gupta [Wed, 4 Mar 2020 05:47:04 +0000 (11:17 +0530)]
net/octeontx: fix meson build for disabled drivers
Add a condition to check if octeontx drivers are disabled.
octeontx drivers are built only if dependent drivers i.e.
ethdev, mempool and common/octeontx are enabled.
Bugzilla ID: 387 Fixes: 7f615033d64f ("drivers/net: build Cavium NIC PMDs with meson") Cc: stable@dpdk.org Signed-off-by: Amit Gupta <agupta3@marvell.com> Reviewed-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Harman Kalra <hkalra@marvell.com>
Igor Romanov [Mon, 30 Mar 2020 10:25:45 +0000 (11:25 +0100)]
net/sfc: check actual all multicast unknown unicast filters
Check that unknown unicast and unknown multicast filters are
applied and return an error if they are not applied. The error
is used in promiscuous and all multicast mode enable and disable
callbacks.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru> Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Igor Romanov [Mon, 30 Mar 2020 10:25:44 +0000 (11:25 +0100)]
net/sfc/base: add API to get currently operating filters
Unknown unicast filter creation may fail because of insufficient
permissions on VF. This failure is handled internally in libefx MAC
reconfiguration without any way for a user to know if it happened.
Making the MAC reconfiguration forward error code of filter
reconfiguration would be too destructive to the existing code
that may rely on the function never returning that error.
Add an API for getting the status of current unknown unicast and
all multicast filters since user must know that requested
filters are actually applied.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru> Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Eduard Serra [Wed, 25 Mar 2020 19:18:51 +0000 (12:18 -0700)]
net/vmxnet3: support MTU set
(Picked up from @Charles Myers patch
https://patchwork.dpdk.org/patch/57771/)
When the mtu_set() function is not implemented, rte_eth_dev_set_mtu()
fails with -ENOTSUP and mtu is not stored in the mtu field in the
rte_eth_dev_data. This causes the mtu in Vmxnet3_MiscConf which is
shared with hypervisor to always be set to 1500.
This may cause issues receiving jumbo frames on Enhanced Data Path
N-VDS.
Signed-off-by: Eduard Serra <eserra@vmware.com> Acked-by: Yong Wang <yongwang@vmware.com>
Guinan Sun [Tue, 24 Mar 2020 04:36:46 +0000 (04:36 +0000)]
net/i40e: enable VF to program MAC address
Due to the restriction of the flag I40E_FLAG_VF_MAC_BY_PF,
VF cannot program the MAC address.
In order to solve this inflexibility, the use of I40E_FLAG_VF_MAC_BY_PF
has been deleted in the code implementation to ensure that
VF can flexibly program the MAC address.
Signed-off-by: Guinan Sun <guinanx.sun@intel.com> Acked-by: Qi Zhang <qi.z.zhang@intel.com> Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
Haiyue Wang [Tue, 31 Mar 2020 06:50:34 +0000 (14:50 +0800)]
net/ice/base: check memory pointer before copying
The ice_memdup doesn't check the new allocated memory pointer, it calls
the rte_memcpy directly. It should check it.
Fixes: 5f0978e96220 ("net/ice/base: add OS specific implementation") Cc: stable@dpdk.org Signed-off-by: Haiyue Wang <haiyue.wang@intel.com> Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
Qi Zhang [Mon, 30 Mar 2020 11:45:38 +0000 (19:45 +0800)]
net/ice/base: enable RSS for PFCP/L2TP/ESP/AH
Add support for PFCP, L2TP, ESP and AH RSS enabling.
Signed-off-by: Ting Xu <ting.xu@intel.com> Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com> Signed-off-by: Qi Zhang <qi.z.zhang@intel.com> Acked-by: Qiming Yang <qiming.yang@intel.com>
Qi Zhang [Mon, 30 Mar 2020 11:45:37 +0000 (19:45 +0800)]
net/ice/base: fix binary order for GTPU filter
Take network order for gtpu fdir filter.
Fixes: b5c274f4e2ad ("net/ice/base: support FDIR for GTPU QFI field") Cc: stable@dpdk.org Signed-off-by: Ting Xu <ting.xu@intel.com> Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com> Signed-off-by: Qi Zhang <qi.z.zhang@intel.com> Acked-by: Qiming Yang <qiming.yang@intel.com>
Qi Zhang [Mon, 30 Mar 2020 11:45:36 +0000 (19:45 +0800)]
net/ice/base: change function to static
Change ice_set_fd_desc_val to static, since it only be used
internally.
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com> Signed-off-by: Qi Zhang <qi.z.zhang@intel.com> Acked-by: Qiming Yang <qiming.yang@intel.com>
Qi Zhang [Mon, 30 Mar 2020 11:45:35 +0000 (19:45 +0800)]
net/ice/base: support PFCP and NAT-T of switch
This patch add support switch rule for PFCP and NAT-T
packet base on profile rule, PFCP and NAT-T packet will not
be matched on any packet fields, but instead matches
the profile that the packet hits in the switch block.
Signed-off-by: Wei Zhao <wei.zhao1@intel.com> Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com> Signed-off-by: Qi Zhang <qi.z.zhang@intel.com> Acked-by: Qiming Yang <qiming.yang@intel.com>
Qi Zhang [Mon, 30 Mar 2020 11:45:34 +0000 (19:45 +0800)]
net/ice/base: move some macros
Move some macro from ice_switch.c to ice_switch.h. Currently this
is only required by kernel driver, DPDK just to sync the code.
Signed-off-by: Xiao Zhang <xiao.zhang@intel.com> Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com> Signed-off-by: Qi Zhang <qi.z.zhang@intel.com> Acked-by: Qiming Yang <qiming.yang@intel.com>
Qi Zhang [Mon, 30 Mar 2020 11:45:33 +0000 (19:45 +0800)]
net/ice/base: group case statements
ICE_BLK_FD and ICE_BLK_RSS are executing the same code so group the case
statements together instead of duplicating code for each block.
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com> Signed-off-by: Qi Zhang <qi.z.zhang@intel.com> Acked-by: Qiming Yang <qiming.yang@intel.com>
Qi Zhang [Mon, 30 Mar 2020 11:45:32 +0000 (19:45 +0800)]
net/ice/base: handle critical FW error
A race condition between FW and SW can occur between admin queue setup
and the first command sent. A link event may occur and FW attempts to
notify a non-existent queue. FW will set the critical error bit and
disable the queue. When this happens retry queue setup.
Signed-off-by: Evan Swanson <evan.swanson@intel.com> Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com> Signed-off-by: Qi Zhang <qi.z.zhang@intel.com> Acked-by: Qiming Yang <qiming.yang@intel.com>
git.droids-corp.org / dpdk.git / log