The current implementation of rte_kni_rx_burst polls the fifo for buffers.
Irrespective of success or failure, it allocates the mbuf and try to put them into the alloc_q
if the buffers are not added to alloc_q, it frees them.
This waste lots of cpu cycles in allocating and freeing the buffers if alloc_q is full.
The logic has been changed to:
1. Initially allocand add buffer(burstsize) to alloc_q
2. Add buffers to alloc_q only when you are pulling out the buffers.
Signed-off-by: Hemant Agrawal <hemant@freescale.com>
Reviewed-by: Jay Rolette <rolette@infiniteio.com>
LPM table overflow may occur if table is full and added rule has
the biggest depth that already have some rules.
Signed-off-by: Igor Ryzhov <iryzhov@nfware.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
* support calling rte_vhost_driver_register after rte_vhost_driver_session_start
* add mutext to protect fdset from concurrent access
* add busy flag in fdentry. this flag is set before cb and cleared after cb is finished.
mutex lock scenario in vhost:
* event_dispatch(in rte_vhost_driver_session_start) runs in a separate thread, infinitely
processing vhost messages through cb(callback).
* event_dispatch acquires the lock, get the cb and its context, mark the busy flag,
and releases the mutex.
* vserver_new_vq_conn cb calls fdset_add, which acquires the mutex and add new fd into fdset.
* vserver_message_handler cb frees data context, marks remove flag to request to delete
connfd(connection fd) from fdset.
* after cb returns, event_dispatch
1. clears busy flag.
2. if there is remove request, call fdset_del, which acquires mutex, checks busy flag, and
removes connfd from fdset.
* rte_vhost_driver_unregister(not implemented) runs in another thread, acquires the mutex,
calls fdset_del to remove fd(listenerfd) from fdset. Then it could free data context.
The above steps ensures fd data context isn't freed when cb is using.
VM(s) should have been shutdown before rte_vhost_driver_unregister.
Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Tetsuya Mukawa <mukawa@igel.co.jp>
for vhost-cuse, ifname is the name of the tap device
for vhost-user, ifname is the name of the unix domain socket path
Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Signed-off-by: Przemyslaw Czesnowicz <przemyslaw.czesnowicz@intel.com>
Acked-by: Tetsuya Mukawa <mukawa@igel.co.jp>
In rte_vhost_driver_register(), vhost unix domain socket listener fd is created
and added to polled(based on select) fdset.
In rte_vhost_driver_session_start(), fds in the fdset are checked for
processing. If there is new connection from qemu, connection fd accepted is
added to polled fdset. The listener and connection fds in the fdset are
then both checked. When there is message on the connection fd, its
callback vserver_message_handler is called to process vhost-user messages.
To support identifying which virtio is from which guest VM, we could call
rte_vhost_driver_register with different socket path. Virtio devices from
same VM will connect to VM specific socket. The socket path information is
stored in the virtio_net structure.
Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Signed-off-by: Przemyslaw Czesnowicz <przemyslaw.czesnowicz@intel.com>
Acked-by: Tetsuya Mukawa <mukawa@igel.co.jp>
remove set_memory_table ops
vhost-cuse or vhost-user will both implement their own set_memory_region handler.
In current vhost-cuse implementation, guest numa memory isn't supported.
Assume that guest memory is backed by only one file.
Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Signed-off-by: Przemyslaw Czesnowicz <przemyslaw.czesnowicz@intel.com>
This functions accepts a virtual address and pid(qemu), and maps it into
current process(vhost)'s address space.
The memory behind the virtual address should be backed by a file,
and virtual address should be the starting address.
Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Acked-by: Tetsuya Mukawa <mukawa@igel.co.jp>
File descriptor is copied from qemu process into vhost process.
vhost-user doesn't need eventfd kernel module to copy fds between processes.
Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Signed-off-by: Przemyslaw Czesnowicz <przemyslaw.czesnowicz@intel.com>
Rename vhost-net-cdev.h to vhost-net.h.
This file defines common operations provided by virtio-net(.c).
Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Acked-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Create vhost_cuse directory and move vhost-net-cdev.c into vhost_cuse.
vhost-cuse driver will be divided into two parts: cuse driver specific message
handling(in cuse directory) and common message handling(in virtio-net.c).
vhost ioctl message is pre-processed in cuse and then sent to virtio-net
if is not terminated.
virtio-net.c provides common message handling for both vhost-cuse and vhost-user.
Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Acked-by: Tetsuya Mukawa <mukawa@igel.co.jp>
VIRTIO_NET_F_CTRL_RX is dependant on VIRTIO_NET_F_CTRL_VQ.
Observed that virtio-net driver in guest would crash with only CTRL_RX enabled.
In virtnet_send_command:
/* Caller should know better */
BUG_ON(!virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ) ||
(out + in > VIRTNET_SEND_COMMAND_SG_MAX));
Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Acked-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Add optional support for inline processing of packets inside the RX
or TX call. For an RX callback, what happens is that we get a set of
packets from the NIC and then pass them to a callback function, if
configured, to allow additional processing to be done on them, e.g.
filling in more mbuf fields, before passing back to the application.
On TX, the packets are similarly post-processed before being handed
to the NIC for transmission.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Signed-off-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
The 'callbacks' member of the rte_eth_dev structure has been renamed
to 'link_intr_cbs' to make it clear that it refers to callbacks from
NIC interrupts. This allows us to add other types of callbacks to
the structure without ambiguity.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Signed-off-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
This patch enables PF's internal switch by setting ALLOWLOOPBACK
flag when VEB is created. With this patch, traffic from PF can be
switched on the VEB.
Test report: http://www.dpdk.org/ml/archives/dev/2015-February/013237.html
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Jijiang Liu <jijiang.liu@intel.com>
Tested-by: Min Cao <min.cao@intel.com>
In i40e_vsi_config_tc_queue_mapping, should add a flag to indicate
another valid setting by OR operation, but not set this flag to
valid_sections, otherwise it will overwrite the flags set before.
Test report: http://www.dpdk.org/ml/archives/dev/2015-February/013237.html
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Jijiang Liu <jijiang.liu@intel.com>
Tested-by: Min Cao <min.cao@intel.com>
On XL710, performance number is far from the expectation on recent
firmware versions, if promiscuous mode is disabled, or promiscuous
mode is enabled and port MAC address is equal to the packet
destination MAC address. The fix for this issue may not be
integrated in the following firmware version. So the workaround in
software driver is needed. For XL710, it needs to modify the initial
values of 3 internal only registers, which are the same as X710.
Note that the values for X710 and XL710 registers could be different,
and the workaround can be removed when it is fixed in firmware in
the future.
Test report: http://www.dpdk.org/ml/archives/dev/2015-February/012749.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
Tested-by: Qian Xu <qian.q.xu@intel.com>
While VFIO doesn't allow us to map complete BARs with MSI-X tables,
it does allow us to map around them in PAGE_SIZE granularity. There
might be adapters that provide their registers in the same BAR
but on a different page. For example, Intel's NVME adapter, though
not a network adapter, provides only one MMIO BAR that contains
the MSI-X table.
Signed-off-by: Dan Aloni <dan@kernelim.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
This patch removes all references to RTE_MBUF_REFCNT, setting the refcnt
field in the mbuf struct permanently.
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Currently for mbufs with refcnt, we cannot free mbufs with external memory
buffers (ie. vhost zero copy), as they are recognized as indirect
attached mbufs and therefore we free the direct mbuf it points to,
resulting in an error in the case of external memory buffers.
We solve the issue by introducing the IND_ATTACHED_MBUF flag, which indicates
that the mbuf is an indirect attached mbuf pointing to another mbuf.
When we free an mbuf, we only free the direct mbuf if the flag is set.
Freeing an mbuf with external buffer is the same as freeing a non attached mbuf.
The flag is set during attach and clear on detach.
So in the case of vhost zero copy where we have mbufs with external
buffers, by default we just free the mbuf and it is up to the user to deal with
the external buffer.
This patch would allow the removal of the RTE_MBUF_REFCNT config option,
setting refcnt for all mbufs permanently.
The patch also modifies the vhost example as it was using the
RTE_MBUF_INDIRECT macro to detect if it was an mbuf with external buffer.
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
This patch introduces CONFIG_RTE_KNI_PREEMPT_DEFAULT flag. When set to 'no',
KNI kernel thread(s) do not call schedule_timeout_interruptible(), which
improves overall KNI performance at the expense of CPU cycles (polling).
Default values is 'yes', maintaining the same behaviour as of now.
Signed-off-by: Marc Sune <marc.sune@bisdn.de>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Calculating hash for data of variable length is more efficient
when that data is sliced into 8-byte pieces. The rest part of data
is hashed using CRC32 functions with either 8 and 4 byte operands.
Signed-off-by: Yerden Zhumabekov <e_zhumabekov@sts.kz>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Initially, SSE4.2 support is detected via the constructor function.
Added rte_hash_crc_set_alg() function to detect and set CRC32
implementation if necessary. SSE4.2 is allowed by default.
rte_hash_crc_*byte() functions reworked so they choose available
CRC32 implementation in the runtime.
Signed-off-by: Yerden Zhumabekov <e_zhumabekov@sts.kz>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
SSE4.2 provides CRC32 intrinsic with 8-byte operand.
Signed-off-by: Yerden Zhumabekov <e_zhumabekov@sts.kz>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Give up using built-in intrinsics and use our own assembly
implementation. Remove #include entry as well.
Signed-off-by: Yerden Zhumabekov <e_zhumabekov@sts.kz>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Added:
- crc32c_sse42_u32() emits 'crc32l' asm instruction;
- crc32c_sse42_u64() emits 'crc32q' asm instruction;
- crc32c_sse42_u64_mimic(), wrapper in case of run on 32-bit platform.
Signed-off-by: Yerden Zhumabekov <e_zhumabekov@sts.kz>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Add lookup tables for CRC32 algorithm, crc32c_1word() and
crc32c_2words() functions returning hash of 32-bit and 64-bit
operand.
Signed-off-by: Yerden Zhumabekov <e_zhumabekov@sts.kz>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
The filter types supported are listed below for NVGRE packet:
1. Inner MAC and Inner VLAN ID.
2. Inner MAC address, inner VLAN ID and tenant ID.
3. Inner MAC and tenant ID.
4. Inner MAC address.
5. Outer MAC address, tenant ID and inner MAC address.
Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Signed-off-by: Declan Doherty <declan.doherty@intel.com>
Add an Ethernet type definition for Transparent Ethernet Bridging.
Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Signed-off-by: Declan Doherty <declan.doherty@intel.com>
RSS offload types were defined separately for 1/10G and 40G NICs,
and have no relationship with flow types. The modifications are to
unify all RSS offload types for all PMDs. Unified RSS offload types
have new and common names which can be used for any PMD or
applications, and decouple from specific hardwares.
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
[Thomas: merge with fm10k]
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Flow types was defined actually for i40e hardware specifically,
and wasn't able to be used for defining RSS offload types of all
PMDs. It removed the enum flow types, and uses macros instead
with new names. The new macros can be used for defining RSS
offload types later. Also modifications are made in i40e and
testpmd accordingly.
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
[Thomas: merge with new flow director API]
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
It wrongly calculates the size of the flow type mask array. The fix
is to align the flow type maximum index ID with the number of
element bit width, and then divide the number of element bit width.
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
The old ethertype filter API was removed in commit 75db20648,
but was still in (newly integrated) version map for ABI.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Following structures are removed:
- rte_2tuple_filter
- rte_5tuple_filter
Following APIs are removed:
- rte_eth_dev_add_2tuple_filter
- rte_eth_dev_remove_2tuple_filter
- rte_eth_dev_get_2tuple_filter
- rte_eth_dev_add_5tuple_filter
- rte_eth_dev_remove_5tuple_filter
- rte_eth_dev_get_5tuple_filter
It also move macros TCP_*_FLAG to rte_eth_ctrl.h, and removes the macro
TCP_UGR_FLAG which is duplicated with TCP_URG_FLAG.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
[Thomas: remove also from version map]
This patch defines new functions dealing with ntuple filters which is
corresponding to 5tuple in HW.
It removes old functions which deal with 5tuple filters.
Ntuple filter is dealt with through entrance ixgbe_dev_filter_ctrl.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
This patch defines new functions dealing with ntuple filters which is
corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple
filter for 82576 in HW.
It removes old functions which deal with 2tuple and 5tuple filters in igb driver.
Ntuple filter is dealt with through entrance eth_igb_filter_ctrl.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
This patch defines ntuple filter type RTE_ETH_FILTER_NTUPLE and its structure rte_eth_ntuple_filter.
It also corrects the typo TCP_UGR_FLAG to TCP_URG_FLAG
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Structure rte_syn_filter is removed.
Following APIs are removed:
- rte_eth_dev_add_syn_filter
- rte_eth_dev_remove_syn_filter
- rte_eth_dev_get_syn_filter
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
[Thomas: remove also from version map]
This patch defines new functions dealing with syn filter.
It removes old functions which deal with syn filter.
Syn filter is dealt with through entrance ixgbe_dev_filter_ctrl.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
This patch defines new functions dealing with syn filter.
It removes old functions of syn filter in igb driver.
Syn filter is dealt with through entrance eth_igb_filter_ctrl.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Structure rte_flex_filter is removed.
Following APIs are removed:
- rte_eth_dev_add_flex_filter
- rte_eth_dev_remove_flex_filter
- rte_eth_dev_get_flex_filter
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
[Thomas: remove also from version map]
This patch defines new functions dealing with flex filter.
It removes old functions of flex filter in igb driver.
Syn filter is dealt with through entrance eth_igb_filter_ctrl.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
This patch implement RTE_ETH_FILTER_FLUSH operation to delete all
flow director filters in ixgbe driver.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
This patch changes the get info operation to be implemented through
filter_ctrl API and RTE_ETH_FILTER_INFO/RTE_ETH_FILTER_STATS ops.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
This patch implement the mask configuration of flow director filter,
by using the mask defined in rte_fdir_conf instead of callback function
fdir_set_masks.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
This patch removes the flexbytes_offset from rte_fdir_conf, because
the flexible payload setting is done by flex_conf instead of flexbytes_offset.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
This patch implement the flexpayload configuration of flow director filter.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
This patch adds RTE_ETH_FLOW_TYPE_RAW and RTE_ETH_RAW_PAYLOAD to support the
flexible payload is started from the beginning of the packet.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
This patch changes the add/delete/update operations to be implemented through
filter_ctrl API and RTE_ETH_FILTER_ADD/RTE_ETH_FILTER_DELETE/RTE_ETH_FILTER_UPDATE ops.
It also removes the callback functions:
- ixgbe_eth_dev_ops.fdir_add_signature_filter
- ixgbe_eth_dev_ops.fdir_update_signature_filter
- ixgbe_eth_dev_ops.fdir_remove_signature_filter
- ixgbe_eth_dev_ops.fdir_add_perfect_filter
- ixgbe_eth_dev_ops.fdir_update_perfect_filter
- ixgbe_eth_dev_ops.fdir_remove_perfect_filter
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
enable/disable interrupt by manipulating a control bit of command
register on NIC's PCIe configuration space.
Signed-off-by: Danny Zhou <danny.zhou@intel.com>
Tested-by: Qun Wan <qun.wan@intel.com>
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Declan Doherty <declan.doherty@intel.com>
Change the EAL PCI code so that it can work with both the
uio_pci_generic in-tree driver, as well as the igb_uio
DPDK-specific driver.
This involves changes to
1) Modify method of retrieving BAR resource mapping information
2) Mapping using resource files in /sys rather than /dev/uio*
2) Setup bus master bit in NIC's PCIe configuration space for
uio_pci_generic.
Signed-off-by: Danny Zhou <danny.zhou@intel.com>
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Declan Doherty <declan.doherty@intel.com>
This patch modify mode older name from
BONDING_MODE_ADAPTIVE_TRANSMIT_LOAD_BALANCING to BONDING_MODE_TLB
This patch also changes order of TEST_ASSERT macro in
test_tlb_verify_slave_link_status_change_failover.
Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com>
Acked-by: Declan Doherty <declan.doherty@intel.com>
This patch add some debug information when using link bonding mode 6.
It prints basic information about ARP packets on RX and TX (MAC, ip,
packet number, arp packet type).
If CONFIG_RTE_LIBRTE_BOND_DEBUG_ALB == y.
If CONFIG_RTE_LIBRTE_BOND_DEBUG_ALB_L1 is enabled instead of previous
one, use show command to see IPv4 balancing from clients.
Signed-off-by: Michal Jastrzebski <michalx.k.jastrzebski@intel.com>
Acked-by: Declan Doherty <declan.doherty@intel.com>
This mode includes adaptive TLB and receive load balancing (RLB). In RLB
the bonding driver intercepts ARP replies send by local system and
overwrites its source MAC address, so that different peers send data to
the server on different slave interfaces. When local system sends ARP
request, it saves IP information from it. When ARP reply from that peer
is received, its MAC is stored, one of slave MACs assigned and ARP reply
send to that peer.
Signed-off-by: Maciej Gajdzica <maciejx.t.gajdzica@intel.com>
Signed-off-by: Michal Jastrzebski <michalx.k.jastrzebski@intel.com>
Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com>
Acked-by: Declan Doherty <declan.doherty@intel.com>
Changed MAC address type from uint8_t[6] to struct ether_addr and IP
address type from uint8_t[4] to uint32_t to make it consistent with
other DPDK code using MAC and IP addresses. It allows us to use
is_same_ether_addr and ether_addr_copy functions on MAC addresses in ARP header. Also
removed union from arp_hdr struct to make calls to arp_data items
shorter. Updated test-pmd to match new arp_hdr version.
Signed-off-by: Maciej Gajdzica <maciejx.t.gajdzica@intel.com>
Acked-by: Declan Doherty <declan.doherty@intel.com>
[Thomas: doxygenize comments]
Remove those hotspots which is unnecessary when early returning occurs.
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
Make virtio not require UIO for some security reasons, this is to match
6WIND's virtio-net-pmd.
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
This makes virtio driver work like ixgbe. Transmit buffers are
held until a transmit threshold is reached. The previous behavior
was to hold mbuf's until the ring entry was reused which caused
more memory usage than needed.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
Need to have do special things to set default mac address.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
Change order of initialization to match Linux kernel.
Don't blow away control queue by doing reset when stopped.
Calling dev_stop then dev_start would not work.
Dev_stop was calling virtio reset and that would clear all queues
and clear all feature negotiation.
Resolved by only doing reset on device removal.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
To keep the consistent logic with normal Rx path, the mergeable
Rx path also needs software vlan strip/decap if it is enabled.
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
Implement VLAN stripping in software. This allows application
to be device independent.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
It is helpful to allow device drivers that don't support hardware
VLAN stripping to emulate this in software. This allows application
to be device independent.
Avoid discarding shared mbufs. Make a copy in rte_vlan_insert() of any
packet to be tagged that has a reference count > 1.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
Since vq_alignment is constant (always 4K), it does not
need to be part of the vring struct.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
Virtio has link state interrupt which can be used.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
Starting driver with link down should be ok, it is with every
other driver. So just allow it.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
Updating the vring descriptor index should be done before notifying host;
Remove 2 duplicated store memory barriers in both Rx and Tx path because there is
store memory barrier in vq_update_avail_idx function;
Notify the host only if packets actually transmitted(nb_tx > 0).
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
The DPDK driver only has to deal with the case of running on PCI
and with SMP. In this case, the code can use the weaker barriers
instead of using hard (fence) barriers. This will help performance.
The rationale is explained in Linux kernel virtio_ring.h.
To make it clearer that this is a virtio thing and not some generic
barrier, prefix the barrier calls with virtio_.
Add missing (and needed) barrier between updating ring data
structure and notifying host.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
Better to check at compile time than fail at runtime.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
Make vtpci_get_status a local function as it is used in one file.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
For clarity make the setup of PCI resources for Linux into a function rather
than block of code #ifdef'd in middle of dev_init.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
Eliminate ambiguity in the condition which trips up a "logical not
is only applied to the left..." warning from gcc 5, causing build
failure with -Werror. Besides non-ambiguous, the condition is
far more obvious this way.
Signed-off-by: Panu Matilainen <pmatilai@redhat.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
CONFIG_RTE_LIBRTE_IXGBE_RX_ALLOW_BULK_ALLOC is a prerequisite
of CONFIG_RTE_IXGBE_INC_VECTOR.
Reported-by: Alexander Belyakov <abelyako@gmail.com>
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
When the vector pmd was receiving a mix of packets of various sizes,
some of which were split across multiple mbufs, there was an issue
with reassembly of the jumbo frames. This was due to a skipped increment
when using "continue" in a while loop. Changing the loop to a "for"
loop fixes this problem, by ensuring the increment is always performed.
Reported-by: Prashant Upadhyaya <praupadhyaya@gmail.com>
Reported-by: Martin Weiser <martin.weiser@allegro-packets.com>
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Tested-by: Martin Weiser <martin.weiser@allegro-packets.com>
ixgbe is little endian, but cpu maybe not.
add necessary conversions.
rte_cpu_to_le_32(...) for PCI write
rte_le_to_cpu_32(...) for PCI read.
Signed-off-by: Xuelin Shi <xuelin.shi@freescale.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
e1000 is little endian, but cpu maybe not.
add necessary conversions.
rte_cpu_to_le_32(...) for PCI write
rte_le_to_cpu_32(...) for PCI read.
Signed-off-by: Xuelin Shi <xuelin.shi@freescale.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Variables are unsigned int but format scans for signed int.
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Building shared libraries and using virtio PMD results in undefined
reference to 'rte_eal_iopl_init'.
Add missing function to eal version map.
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
The argument ressize contains the size of the result buffer which
should be large enough to store the parsed result of a token. In
this case, it should be larger or equal to sizeof(cmdline_portlist_t)
(4 bytes), not PORTLIST_TOKEN_SIZE which is the max size of the token
string.
This is not a critical, it fixes cases where the total length of the
parsed instruction is greater than the maximum.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
rte_eth_bond_8023ad_setup and rte_eth_bond_8023ad_conf_get entries need
to be exported to be used by test application for shared libraries
compilation.
Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Acked-by: Declan Doherty <declan.doherty@intel.com>
Link up and down implementation.
Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Acked-by: Declan Doherty <declan.doherty@intel.com>
rte_eth_stats_get is to get physical device statistics. Without
return status, caller does not know whether function fails or not
(i.e. invalid port_id, driver has no stats_get callback). Caller
cannot differiente normal 0 stats or error case. This fix adds
a return status to the function.
Signed-off-by: Jia Yu <jyu@vmware.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Since LSR interrupt is disabled by pmd drivers, link status in rte_eth_device is always down.
Bond slave_configure() enables LSR interrupt on devices to get notification if link status
changes. However, the LSC interrupt at device start time is still lost.
In this fix, call link_update to read link status from hardware
register at device start time.
Signed-off-by: Jia Yu <jyu@vmware.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
This library provides reordering capability for out of order mbufs based
on a sequence number in the mbuf structure.
Signed-off-by: Reshma Pattan <reshma.pattan@intel.com>
Signed-off-by: Richardson Bruce <bruce.richardson@intel.com>
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Declan Doherty <declan.doherty@intel.com>
As far as I know, there is no reason why we should have a limit on the length of
parameters that can be given for a device.
Remove this limit by using dynamic allocations.
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Prepare for next commit.
Fix some indent issues, refactor error code.
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
1. Add functions to enable PF/VF interrupt.
2. Add function to process error message passed from interrupt.
2. Add 2 interrupt handling functions, one for PF and one for VF.
2. Enable interrupt after completing initialization of NIC.
Signed-off-by: Jeff Shaw <jeffrey.b.shaw@intel.com>
Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
fm10k pmd driver will support both PF and VF device with single
copy of code. The reason is NIC maps registers with same
function in PF and VF to same PCI I/O address. Then, PF/VF drivers
use same address to access registers belonging to it, HW will
translate the request to correct units.
For some functionalities that are unique to PF, driver will check
current driver type and behave correctly.
Signed-off-by: Jeff Shaw <jeffrey.b.shaw@intel.com>
Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
1. Add fm10k_recv_scattered_pkts function to receive jumbo frame
and multi-segment packets.
2. Configure correct receive function in rx_init and dev_init.
Signed-off-by: Jeff Shaw <jeffrey.b.shaw@intel.com>
Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
1. Configure RSS in fm10k_dev_rx_init function.
2. Add fm10k_rss_hash_update and fm10k_rss_hash_conf_get to get
and inquery RSS configuration.
Signed-off-by: Jeff Shaw <jeffrey.b.shaw@intel.com>
Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
1. Add fm10k_recv_pkts and fm10k_xmit_pkts functions.
2. Link app function pointer to actual fm10k recv/xmit
functions.
3. Change Makefile to compile new file fm10k_rxtx.c
Signed-off-by: Jeff Shaw <jeffrey.b.shaw@intel.com>
Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
1. Add function to initialize RX queues.
2. Add function to initialize TX queues.
3. Add fm10k_dev_start, fm10k_dev_stop and fm10k_dev_close
functions.
4. Add function to close mailbox service.
Signed-off-by: Jeff Shaw <jeffrey.b.shaw@intel.com>
Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
1. Add 4 functions fm10k_dev_rx_queue_start,
fm10k_dev_rx_queue_stop, fm10k_dev_tx_queue_start,
and fm10k_dev_tx_queue_stop.
2. verify Rx packet buffer alignment is valid.
Hardware requires specific alignment for Rx packet buffers. At
least one of the following two conditions must be satisfied.
1) Address is 512B aligned
2) Address is 8B aligned and buffer does not cross 4K boundary.
Alignment is checked by the driver when the Rx queue is reset. It
is assumed that if an entire descriptor ring can be filled with
buffers containing valid alignment, then all buffers in that mempool
have valid address alignment. It is the responsibility of the user
to ensure all buffers have valid alignment, as it is the user who
creates the mempool.
It is assumed the buffer needs only to store a maximum size Ethernet
frame.
Signed-off-by: Jeff Shaw <jeffrey.b.shaw@intel.com>
Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
1. Add init function to scan and initialize fm10k PF device.
2. Add implementation to register fm10k pmd PF driver.
3. Add 3 functions fm10k_dev_configure, fm10k_stats_get and
fm10k_stats_get.
4. Add fm10k.h to define macros and basic data structure.
5. Add fm10k_logs.h to control log message output.
6. Change config/common_bsdapp and config/common_linuxapp, add
macros to control fm10k pmd driver compile for linux and bsd.
7. Add Makefile.
8. Change lib/Makefile to add fm10k driver into compile list.
9. Change mk/rte.app.mk to add fm10k lib into link.
10. Add ABI version of librte_pmd_fm10k
Signed-off-by: Jeff Shaw <jeffrey.b.shaw@intel.com>
Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
Signed-off-by: Michael Qiu <michael.qiu@intel.com>
Add fm10k device ID list into rte_pci_dev_ids.h.
Signed-off-by: Jeff Shaw <jeffrey.b.shaw@intel.com>
Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
Base driver is developed and maintained by Intel ND team, includes
basic functional service to Intel Ethernet Switch FM10000 Series
of silicons.
Any suggestion on bug fix and improvement within this directory is
welcome, but need this team to change and update.
Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com>
This could be useful to have this values for debug purposes.
Suggested-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Jijiang Liu <jijiang.liu@intel.com>
When offloading the checksums of ipip tunnels, m->l2_len is set to 0
as there is no tunnel or inner l2 header. Since this is a valid value
remove the test.
By the way, also remove the same test with l3_len because at this
point, it is expected that the software provides proper values in the
mbuf. It should avoid a test in dataplane processing and therefore
slightly increase performance.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Jijiang Liu <jijiang.liu@intel.com>
Advertise the DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM flag in the PMD
features. It means that the i40e PMD supports the offload of outer IP
checksum when transmitting tunneling packet.
Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
If the flag is advertised by a PMD, the NIC supports the outer IP
checksum TX offload of tunneling packets, therefore an application can
set the PKT_TX_OUTER_IP_CKSUM flag in mbufs when transmitting on this
port.
Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Since previous commit, the flag PKT_TX_UDP_TUNNEL_PKT is not used by any PMD,
remove it from mbuf API and from csumonly (testpmd). In csumonly, the
PKT_TX_OUTER_IP_CKSUM flag is already set for vxlan checksum, providing
enough information to the underlying driver.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Jijiang Liu <jijiang.liu@intel.com>
The definition of the flag PKT_TX_UDP_TUNNEL_PKT in rte_mbuf.h was:
TX packet is an UDP tunneled packet. It must be specified when using
outer checksum offload (PKT_TX_OUTER_IP_CKSUM)
This flag was used to tell the NIC that the offload type is UDP
(I40E_TXD_CTX_UDP_TUNNELING flag). In the datasheet, it says it's
required to specify the tunnel type in the register. However, some tests
(see [1]) showed that it also works without this flag.
Moreover, it is not explained how the hardware use this
information. From a network perspective, this information is useless for
calculating the outer IP checksum as it does not depend on the payload.
Having this flag in the API would force the application to specify the
tunnel type for something that looks only useful for this PMD. It will
limit the number of possible tunnel types (we would need a flag for each
tunnel type) and therefore prevent to support outer IP checksum for
proprietary tunnels.
Finally, if a hardware advertises "I support outer IP checksum", it must
be supported for any payload types.
This has been validated by [2], knowing that the ipip test case was fixed
after this test report [3].
[1] http://dpdk.org/ml/archives/dev/2015-January/011380.html
[2] http://dpdk.org/ml/archives/dev/2015-January/011475.html
[3] http://dpdk.org/ml/archives/dev/2015-January/011610.html
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Jijiang Liu <jijiang.liu@intel.com>
From i40e datasheet:
The IP header type and its offload. In case of tunneling, the IIPT
relates to the inner IP header. See also EIPT field for the outer
(External) IP header offload.
00 - non IP packet or packet type is not defined by software
01 - IPv6 packet
10 - IPv4 packet with no IP checksum offload
11 - IPv4 packet with IP checksum offload
Therefore it is not needed to fill the IIPT field if no offload is
requested (we can keep the value to 00). For instance, the linux driver
code does not set it when (skb->ip_summed != CHECKSUM_PARTIAL). We can
do the same in the dpdk driver.
The function i40e_txd_enable_checksum() that fills the offload registers
can only be called for packets requiring an offload.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Jijiang Liu <jijiang.liu@intel.com>
The alias PKT_TX_IPV4_CSUM is only used in one place of i40e driver.
Remove it and only keep the legacy flag PKT_TX_IP_CSUM.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Jijiang Liu <jijiang.liu@intel.com>
In loopback mode, it's expected force link up even when there's no cable connect.
But in codes, setup_sfp() rewrites the related register.
It causes in the case 'multispeed_fiber', it can't link up without cable connect.
Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Patrick Lu <patrick.lu@intel.com>
max_vfs will only be created by igb_uio driver, for other
drivers like vfio or pci_uio_generic, max_vfs will miss.
But sriov_numvfs is not driver related, just get the vf numbers
from that field.
Signed-off-by: Michael Qiu <michael.qiu@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
cmdline_token_portlist_ops fell through cracks in the initial symbol
versioning patch, breaking pktgen build.
Signed-off-by: Panu Matilainen <pmatilai@redhat.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
To differentiate libraries that break ABI, we add a library version number
suffix to the library, which must be incremented when a given libraries ABI is
broken. This patch enforces that addition, sets the initial abi soname
extension to 1 for each library and creates a symlink to the base SONAME so that
the test applications will link properly.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Add linker version script files to each DPDK library to put a stake in the
ground from which we can start cleaning up API's
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Add initial pass header files to support symbol versioning.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Hash filter control has been implemented for i40e. It includes
getting/setting,
- global hash configurations (hash function type, and symmetric
hash enable per flow type)
- symmetric hash enable per port
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
In order to support hash filter configuration, filter type of hash
is added, also the corresponding structures, macros and definitions
are added.
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Calculating the default RSS hash keys at run time is not needed
at all, and may have race conditions. The alternative is to use
array of random values which were generated manually as the
default hash keys.
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
The Link bonding library is incorrectly using receive packet type flags
in the transmit policy hashing functions, which would cause packets
generated locally to be incorrectly distributed across the slave
devices. This patch completely removes the dependency on the packet
type flags and uses the ether_type from either the Ethernet header or
the VLAN headers for branching.
This patch also includes the associate changes in the test suite and in
the packet_burst_generator code to remove the dependences on the packet
type flags.
Signed-off-by: Declan Doherty <declan.doherty@intel.com>
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
This is a duplication of some EAL parts for a standalone packaging
which is not documented.
Packaging should be done outside of DPDK.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
If at build phase we don't make any trie splitting,
then temporary build structures and resulting RT structure might be
much bigger than current.
>From other side - having just one trie instead of multiple can speedup
search quite significantly.
>From my measurements on rule-sets with ~10K rules:
RT table up to 8 times bigger, classify() up to 80% faster
than current implementation.
To make it possible for the user to decide about performance/space trade-off -
new parameter for build config structure (max_size) is introduced.
Setting it to the value greater than zero, instructs rte_acl_build() to:
- make sure that size of RT table wouldn't exceed given value.
- attempt to minimise number of tries in the table.
Setting it to zero maintains current behaviour.
That introduces a minor change in the public API, but I think the possible
performance gain is too big to ignore it.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Vector code reorganisation/deduplication:
To avoid maintaining two nearly identical implementations of calc_addr()
(one for SSE, another for AVX2), replace it with a new macro that suits
both SSE and AVX2 code-paths.
Also remove no needed any more MM_* macros.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Reorganise SSE code-path a bit by moving lo/hi dwords shuffle
out from calc_addr().
That allows to make calc_addr() for SSE and AVX2 practically identical
and opens opportunity for further code deduplication.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Previous improvements made scalar method the fastest one
for tiny bunch of packets (< 4).
That allows us to remove specific vector code-path for small number of packets
(search_sse_2) and always use scalar method for such cases.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Introduce new classify() method that uses AVX2 instructions.
>From my measurements:
On HSW boards when processing >= 16 packets per call,
AVX2 method outperforms it's SSE counterpart by 10-25%,
(depending on the ruleset).
When build with the compilers that don't support AVX2 instructions,
make rte_acl_classify_avx2() do nothing and return an error.
At runtime, if librte_acl was build with the compiler that supports AVX2,
this method is selected as default one on HW that supports AVX2.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
New data type to manipulate 256 bit AVX values.
Rename field in the rte_xmm to keep common naming across SSE/AVX fields.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Move common check for input parameters up into rte_acl_classify_alg().
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Make classify_scalar to behave in the same way as it's vector counterpart:
move match check out of the inner loop, etc.
That makes scalar and vector code look more identical.
Plus it improves scalar code performance.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Right now we allocate indexes for all types of nodes, except MATCH,
at 'gen final RT table' stage.
For MATCH type nodes we are doing it at building temporary tree stage.
This is totally unnecessary and makes code more complex and error prone.
Rework the code and make MATCH indexes being allocated at the same stage
as all others.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Introduced division of whole 256 child transition enties
into 4 sub-groups (64 kids per group).
So 2 groups within the same node with identical children,
can use one set of transition entries.
That allows to compact some DFA nodes and get space savings in the RT table,
without any negative performance impact.
>From what I've seen an average space savings: ~20%.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
There was a bug at build phase that can cause matches beeing overwritten.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Current rule-wildness based heuristics can cause unnecessary splits of
the ruleset.
That might have negative performance effect:
more tries to traverse, bigger RT tables.
After removing it, on some test-cases with big rulesets (~10K)
observed ~50% speedup.
No difference for smaller rulesets.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Make data_indexes long enough to survive idle transitions.
That allows to simplify match processing code.
Also fix incorrect size calculations for data indexes.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Fix compilation with RTE_LIBRTE_ACL_STANDALONE=y
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
For ixgbe, when queue start fails, for example, mbuf allocate
failure, the device will still start successfully, which could be
an issue.
Add return status check of queue start to avoid this issue.
Signed-off-by: Michael Qiu <michael.qiu@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
To follow up the comments from Pawel Wodkowski, remove this unnecessary check,
as check_mq_mode has already check the queue number in device configure stage,
if the queue number of vf is not correct, it will return error code and exit,
so it doesn't need check again here in device start stage (note: pf_host_configure
is called in device start stage).
Fixes: 42d2f78abc ("configure VF RSS")
Suggested-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
The link_status variable is not set when device is initialized.
This can lead to problems with link never being reported as up
if using some SFP modules where the link is instantly on.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
The rte_eth_stats_get is the only API that should call the device
statistics function directly, and it already does a memset of the
resulting structure since commit 02331c16ec. Therefore doing
memset() in the driver is redundant and should be removed.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
[David: remove also in igbvf and pcap PMDs]
Acked-By: David Marchand <david.marchand@6wind.com>
Add missing extern 'C' decls in rte_ip_frag.h.
Fixes: 601e279df0 ("move fragmentation/reassembly headers into a library")
Signed-off-by: Marc Sune <marc.sune@bisdn.de>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
rte_power_freq_min function did not include "extern" keyword,
causing linking errors.
Fixes: 445c6528b5 ("power: common interface for guest and host")
Reported-by: Ildar Mustafin <imustafin@bk.ru>
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Ethernet device's data should contain the virtual device name for pcap port.
This name is correctly set by rte_eth_dev_allocate() at initialization time,
but it is directly lost.
Fixes: 83b4113693 ("ethdev: add unique name to devices")
Signed-off-by: Remi Pommarel <repk@triplefau.lt>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
When using core list argument to define which core to enable (ie -l) the
core_num field of the rte configuration is not updated the same way as using
coremask. This causes rte_lcore_num() to yield different value from the one
using coremask.
Fixes: d888cb8b96 ("add core list input format")
Signed-off-by: Remi Pommarel <repk@triplefau.lt>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Structure rte_ethertype_filter is removed.
Following APIs are removed:
- rte_eth_dev_add_ethertype_filter
- rte_eth_dev_remove_ethertype_filter
- rte_eth_dev_get_ethertype_filter
It is replaced by filter_ctrl API.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
This patch removes old functions which deal with ethertype filter in ixgbe driver.
It also defines ixgbe_dev_filter_ctrl which is binding to filter_ctrl API,
and ethertype filter can be dealt with through this new entrance.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
This patch removes old functions which deal with ethertype filter in igb driver.
It also defines eth_igb_filter_ctrl which is binding to filter_ctrl API,
and ethertype filter can be dealt with through this new entrance.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
In commit 2fc8d6d the behaviour of function rte_is_power_of_2 was
changed to not return true for 0. memzone_reserve_aligned_thread_unsafe
and rte_malloc_socket both make the assumption that for align = 0
!rte_is_power_of_2(align) will return false. This patch adds a check
that align parameter is non-zero before doing the power of 2 check.
Signed-off-by: Declan Doherty <declan.doherty@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
[Thomas: use && operator instead of ternary ?: and fix precedence with parens]
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
It needs config RSS and IXGBE_MRQC and IXGBE_VFPSRTYPE to enable VF RSS.
The psrtype will determine how many queues the received packets will distribute to,
and the value of psrtype should depends on both facet: max VF rxq number which
has been negotiated with PF, and the number of rxq specified in config on guest.
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Reviewed-by: Vlad Zolotarov <vladz@cloudius-systems.com>
Check mq mode for VMDq RSS, handle it correctly instead of returning an error;
Also remove the limitation of per pool queue number has max value of 1, because
the per pool queue number could be 2 or 4 if it is VMDq RSS mode;
The number of rxq specified in config will determine the mq mode for VMDq RSS.
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Reviewed-by: Vlad Zolotarov <vladz@cloudius-systems.com>
Get the available Rx and Tx queue number when receiving IXGBE_VF_GET_QUEUES
message from VF.
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Reviewed-by: Vlad Zolotarov <vladz@cloudius-systems.com>
Negotiate API version with VF when receiving the IXGBE_VF_API_NEGOTIATE message.
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Reviewed-by: Vlad Zolotarov <vladz@cloudius-systems.com>
Put global register configuring out of loop for queue; also fix typo and indent.
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Reviewed-by: Vlad Zolotarov <vladz@cloudius-systems.com>
The lack of result checking of fscanf function, breaks compilation
for default "-Werror=unused-result" flag.
Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Replace d_thread_t with struct thread in nic_uio.
Ref: https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=196691
Quote:
"The d_thread_t typedef is a compat shim to support FreeBSD 4.x.
I'm planning to remove this shim from 11 and dpdk is very unlikely
to ever be ported to 4.x.
If it does it will need far more changes than just d_thread_t"
Reported-by: John Baldwin <jhb@freebsd.org>
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
This patch contains a fix for link bonding handling of vlan tagged packets in mode 3 and 5.
Currently xmit_slave_hash function misinterprets the PKT_RX_VLAN_PKT flag to mean that
there is a vlan tag within the packet when in actually means that there is a valid entry
in the vlan_tci field in the mbuf.
- Fixed VLAN tag support in hashing functions.
- Adds support for TCP in layer 4 header hashing.
- Splits transmit hashing function into separate functions for each policy to
reduce branching and to make the code clearer.
- Fixed incorrect flag set in test application packet generator.
Test report: http://dpdk.org/ml/archives/dev/2015-January/010792.html
Signed-off-by: Declan Doherty <declan.doherty@intel.com>
Acked-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Tested-by: SunX Jiajia <sunx.jiajia@intel.com>
When vfio module is not loaded when kernel support vfio feature,
the routine still try to open the container to get file
description.
This action is not safe, and of course got error messages:
EAL: Detected 40 lcore(s)
EAL: unsupported IOMMU type!
EAL: VFIO support could not be initialized
EAL: Setting up memory...
This may make user confuse, this patch make it reasonable
and much more smooth to user.
Signed-off-by: Michael Qiu <michael.qiu@intel.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
The read/seek/close stub functions are unnecessary on the
log stream. Per glibc fopencookie man page:
cookie_read_function_t *read
If *read is a null pointer, then reads from the custom stream
always return end of file.
cookie_seek_function_t *seek
If *seek is a null pointer, then it is not possible to perform
seek operations on the stream.
cookie_close_function_t *close
If *close is NULL, then no special action is performed when the
stream is closed.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
When scanning the hugetlbfs maps search only for the DPDK maps.
This will allow the application create its own hugetlbfs mappings
and use the DPDK facilities on the same hugetlbfs mount point.
Signed-off-by: Vlad Zolotarov <vladz@cloudius-systems.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
rte_is_power_of_2 returns true for 0 and 0 is not power_of_2.
Fix by checking for n.
Signed-off-by: Ravi Kerur <rkerur@gmail.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
This patch fixes checking the link state of a virtual function. If the
state has already been checked, it does not need to be checked
again. Previously, get_link_status in the ixgbe_hw struct was used to
track if the information had already been retrieved, but this field
was always set to false (signifying that the information was
up-to-date). The problem was introduced by commit 8ef32003 which was
part of a patch set to update the ixgbe portion of the PMD. This patch
does not break consistency with the ixgbevf driver. Instead, it fixes
the problem at the level of DPDK.
Applications that rely on the reported link speed could fail without
this patch. The qos_sched example application provided with DPDK did
not run when virtual functions were used. The output for this example
application is shown below:
EAL: Error - exiting with code: 1
Cause: Unable to config sched subport 0, err=-2
The problem and the effect of the patch can been seen by running the
l2fwd example application using the following command:
sudo ./build/l2fwd -c 0x3 -n 4 -- -p 0x3 -T 0
Before the patch has been applied (with both links up):
...
Checking link statusdone
Port 0 Link Up - speed 100 Mbps - half-duplex
Port 1 Link Up - speed 100 Mbps - half-duplex
L2FWD: entering main loop on lcore 1
...
After the patch has been applied (with both links up):
...
Checking link statusdone
Port 0 Link Up - speed 10000 Mbps - full-duplex
Port 1 Link Up - speed 10000 Mbps - full-duplex
L2FWD: entering main loop on lcore 1
...
Before the patch has been applied (with link 0 down, link 1 up):
...
Checking link statusdone
Port 0 Link Up - speed 100 Mbps - half-duplex
Port 1 Link Up - speed 100 Mbps - half-duplex
L2FWD: entering main loop on lcore 1
...
After the patch has been applied (with link 0 down, link 1 up):
...
Checking link status............................................................
..............................done
Port 0 Link Down
Port 1 Link Up - speed 10000 Mbps - full-duplex
...
Signed-off-by: Balazs Nemeth <balazs.nemeth@intel.com>
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
EAL: probe driver: 8086:10fb rte_ixgbe_pmd
EAL: PCI memory mapped at 0x7f18c2a00000
EAL: PCI memory mapped at 0x7f18c2a80000
Segmentation fault (core dumped)
This is introduced by commit: 46bc9d75
ixgbe: fix multi-process support
When start primary process with command line:
./app/test/test -n 1 -c ffff -m 64
then start the second one:
./app/test/test -n 1 --proc-type=secondary --file-prefix=rte
This segment-fault will occur.
Root cause is test app on primary process only starts device, but
the queue need initialized by manually command line.
So the tx queue is still NULL when secondary process startup.
Reported-by: Yong Liu <yong.liu@intel.com>
Signed-off-by: Michael Qiu <michael.qiu@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
This patch removes the interrupt registration code which was under the flag
VFIO_PRESENT and relies on the rte_lib code for the same.
This also ignores the initial trigger of ISR from the lib.
Signed-off-by: Sujith Sankar <ssujith@cisco.com>
In rte_pmd_init_internals, we are mapping memory but not released
if error occurs it could produce memory leak.
Add unmmap function to release memory.
Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com>
Acked-by: Michal Jastrzebski <michalx.k.jastrzebski@intel.com>
Acked-by: John W. Linville <linville@tuxdriver.com>
In rte_eth_af_packet.c we are we are missing NULL pointer
checks after calls to allocate memory for queues.
Add checking NULL pointer and error handling.
Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Back in commit aaa662e75c ("cmdline: fix overflow on bsd"),
the author failed to fixup a call to cmdline_parse_etheraddr in xenvirt.
This patch makes the needed correction to avoid a build break.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
This patch fixes the issue whereby when using userspace vhost ports
in the context of vSwitching, the name provided to the hypervisor/QEMU
of the vhost tap device needs to be exposed in the library, in order
for the vSwitch to be able to direct packets to the correct device.
This patch introduces an 'ifname' member to the virtio-net structure
which is populated with the tap device name when QEMU is brought up
with a vhost device.
Signed-off-by: Ciara Loftus <ciara.loftus@intel.com>
Signed-off-by: Anthony Fee <anthonyx.fee@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
From CentOS 6.6, function skb_set_hash is introduced, this breaks
the previous assumption. So modify RHEL_RELEASE_VERSION from 7.0
to 6.6 to fix build for rte_kni.ko.
Related mail from Barak Enat:
http://dpdk.org/ml/archives/dev/2014-December/010124.html
building error likes:
CC [M] lib/librte_eal/linuxapp/kni/e1000_82575.o
In file included from lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_osdep.h:41,
from lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_hw.h:31,
from lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_api.h:31,
from lib/librte_eal/linuxapp/kni/e1000_82575.c:38:
lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h:3870: error: conflicting types for ‘skb_set_hash’
include/linux/skbuff.h:620: note: previous definition of ‘skb_set_hash’ was here
Reported-by: Barak Enat <barak@saguna.net>
Signed-off-by: Jincheng Miao <jincheng.miao@gmail.com>
Compile warning which is treated as error occurs on Oracle Linux
(kernel 2.6.39, gcc 4.4.7) as below, or RHEL, CentOS. Aliasing
'struct i40e_aqc_debug_reg_read_write' should be avoided. Use the
elements inside that structure directly can fix the issue.
lib/librte_pmd_i40e/i40e_ethdev.c: In function 'eth_i40e_dev_init':
lib/librte_pmd_i40e/i40e_ethdev.c:5318: error: dereferencing pointer
'cmd' does break strict-aliasing rules
lib/librte_pmd_i40e/i40e_ethdev.c:5314: note: initialized from here
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Seen on RHEL-6.5:
lib/librte_eal/linuxapp/kni/kni_vhost.c:222:
error: ‘struct socket’ has no member named ‘wq’
lib/librte_eal/linuxapp/kni/kni_vhost.c:313:
error: implicit declaration of function ‘sk_sleep’
lib/librte_eal/linuxapp/kni/kni_vhost.c:313:
error: passing argument 1 of ‘__wake_up’ makes pointer from integer without a cast
include/linux/wait.h:146: note: expected ‘struct wait_queue_head_t *’
but argument is of type ‘int’
lib/librte_eal/linuxapp/kni/kni_vhost.c:580:
error: assignment makes pointer from integer without a cast
RHEL6.5 kernel is based on 2.6.32. But there are two changing
from 2.6.35:
1. socket struct is changed
It wrappered previous wait_queue_head_t of socket to
struct socket_wq. So for the kernel older than 2.6.35, we should
directly use socket->wait instead.
2. new function sk_sleep()
This function is implemented from 2.6.35 to obtain wait queue
from struct sock. This patch adds a macro in kni/compat.h
to be compatible with older kernels.
Patch is tested in RHEL6.5 and RHEL7.0 with:
CONFIG_RTE_LIBRTE_KNI=y
CONFIG_RTE_KNI_KO_DEBUG=y
CONFIG_RTE_KNI_VHOST=y
CONFIG_RTE_KNI_VHOST_MAX_CACHE_SIZE=1024
CONFIG_RTE_KNI_VHOST_VNET_HDR_EN=y
CONFIG_RTE_KNI_VHOST_DEBUG_RX=y
CONFIG_RTE_KNI_VHOST_DEBUG_TX=y
Signed-off-by: Jincheng Miao <jmiao@redhat.com>
In commit 59d0ecdbf0 ("MTU accessors"),
max_frame_size was replaced with mtu.
Default size is ETHER_MTU = 1500.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
The cleanup code on error checks for *internals being NULL only after
using the pointer to perform other cleanup. Fix this by moving the
clean-up based on the pointer inside the check for NULL.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Since commit fbde27f19a "get default Rx/Tx configuration from dev info",
a default RX/TX configuration can be used for all PMDs.
In case of vmxnet3, the whole structure was zeroed and not filled out.
The PMD does not support multi segments or offload functions,
so txq_flags should have those flags set.
Test report: http://dpdk.org/ml/archives/dev/2014-December/009933.html
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Tested-by: Xiaonan Zhang <xiaonanx.zhang@intel.com>
On X710, performance number is far from the expectation on recent
firmware versions. The fix for this issue may not be integrated in
the following firmware version. So the workaround in software driver
is needed. It needs to modify the initial values of 3 internal only
registers. Note that the workaround can be removed when it is fixed
in firmware in the future.
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jing Chen <jing.d.chen@intel.com>
Add missing setup for X540 MAC type when setting up VF.
Additional check exists in Linux driver but not in DPDK.
Signed-off-by: Bill Hong <bhong@brocade.com>
Signed-off-by: Stephen Hemminger <shemming@brocade.com>
When using multiple processes, the TX function used in all processes
should be the same, otherwise the secondary processes cannot transmit
more than tx-ring-size - 1 packets.
To achieve this, we extract out the code to select the ixgbe TX function
to be used into a separate function inside the ixgbe driver, and call
that from a secondary process when it is attaching to an
already-configured NIC.
Testing with symmetric MP app shows that we are able to RX and TX from
both primary and secondary processes once this patch is applied.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Reshma Pattan <reshma.pattan@intel.com>
Switch the order of the conditions in a while loop, so we check the
range of "i" against the max, before using it to index into the array.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
This patch fixes build failing with undefined symbol _PAGE_IOMAP with
kernel 3.18.
The Xen-specific _PAGE_IOMAP PTE flag was removed in kernel 3.18 and
could be used for other purpose in future. This patch ensures that
_PAGE_IOMAP flag is only used for kernels before 3.18.
Signed-off-by: Shu Shen <shu.shen@radisys.com>
Acked-by: Jincheng Miao <jmiao@redhat.com>
Enqueue and dequeue burst functions always return a positive
value (including 0), so return type should be unsigned,
instead of int.
Fixed also API doc for one of the functions.
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
For rte_ipv6_phdr_cksum() gcc 4.8.* with "-O3" not always generates
correct code.
Sometimes it 'forgets' to put len and proto fields of psd_header on the stack.
To overcome that problem and speedup things a bit, refactored rte_raw_cksum()
by splitting ipv6 pseudo-header csum calculation into 3 phases:
1. calc sum for src & dst addresses
2. add sum for proto & len.
3. finalise sum
That makes gcc to generate valid code and helps to avoid any copying.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
If the file to be read by the cfgfile is empty, i.e. no configuration
data, but possibly comments present, the cfgfile should not mark the
last processed section (curr_section) as having N entries, since there
is no last processed section.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
When printing the version string to a local variable, use snprintf for
safety over sprintf. This is general good practice even if the values
to print are all hard-coded.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
GCC 4.5.1 from SUSE throws this error:
lib/librte_pmd_enic/enic_main.c:862:2: error: value computed is not used
This change use statements in expressions C extension provided by gcc to avoid
'value computed is not used' warning/error when size is not known at compile
time.
Reported-by: Michael Qiu <michael.qiu@intel.com>
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Acked-by: Michael Qiu <michael.qiu@intel.com>
[Thomas: apply same fix to ppc_64]
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Compile warnings/errors was found on gcc 4.7.2 as follows. Variables
was reported of being used but uninitialized. Assigning an initial
value to it is needed.
lib/librte_pmd_enic/vnic/vnic_dev.c: In function vnic_dev_get_mac_addr:
lib/librte_pmd_enic/vnic/vnic_dev.c:393:16: error: a1 may be used uninitialized
in this function [-Werror=uninitialized]
lib/librte_pmd_enic/vnic/vnic_dev.c:629:10: note: a1 was declared here
lib/librte_pmd_enic/vnic/vnic_dev.c: In function vnic_dev_set_mac_addr:
lib/librte_pmd_enic/vnic/vnic_dev.c:393:16: error: a1 may be used uninitialized
in this function [-Werror=uninitialized]
lib/librte_pmd_enic/vnic/vnic_dev.c:980:10: note: a1 was declared here
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Adding call to mac_address_slaves_update from the lsc handler when the
first slave become active to propagate any mac changes made while
devices are inactive
Changed removing slave logic to use memmove instead of memcpy to move
data within the same array, as this was corrupting the slave array.
Adding unit test to cover failing assignment scenarios
Test report: http://dpdk.org/ml/archives/dev/2014-December/009623.html
Signed-off-by: Declan Doherty <declan.doherty@intel.com>
Acked-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Tested-by: SunX Jiajia <sunx.jiajia@intel.com>
From upstream kernel commit 3db2e9cd, strict_strto* serial functions
are removed. So that we should directly used kstrtoul instead.
Add xen_dom0/compat.h to be compatible with older kernel.
Signed-off-by: Jincheng Miao <jmiao@redhat.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
From upstream kernel commit 3db2e9cd, strict_strto* serial functions
are removed. So that we should directly used kstrtoul instead.
Add kni/compat.h to be compatible with older kernel.
Signed-off-by: Jincheng Miao <jmiao@redhat.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
From upstream kernel commit 3db2e9cd, strict_strto* serial functions
are removed. So that we should directly used kstrtoul instead.
kstrtoul exists from RHEL6.4, so for compatibility with old kernel and RHEL,
add some logic to igb_uio/compat.h.
Signed-off-by: Jincheng Miao <jmiao@redhat.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Function pci_num_vf() is introduced from upstream linux-2.6.34. So
this patch make compatible with longterm kernel linux-2.6.32.63.
For RHEL, function pci_num_vf() begins from RHEL5 update9. And
it is stub-defined when CONFIG_PCI_IOV is not enabled.
So dropped the CONFIG_PCI_IOV checking of commit 11ba0426.
For other distro like RHEL behaved to pci_num_vf(), we could simply
append following condition macro:
(!(defined(OTHER_RELEASE_CODE) && \
OTHER_RELEASE_CODE >= OTHER_RELEASE_VERSION(X, Y)))
Signed-off-by: Jincheng Miao <jmiao@redhat.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
The name of the rte_eth_fdir_flow's rte_eth_ipv6_flow attribute,
'ip6_flow', clashes with a macro defined in
/usr/include/netinet/ip6.h, such that when DPDK is linked with an
application that uses the afforementioned header, the macro is
expanded within the DPDK struct, causing a compilation error.
Rename the relevant attribute in DPDK to resolve this.
Signed-off-by: Mark Kavanagh <mark.b.kavanagh@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
include/rte_ip.h:161: error: dereferencing pointer ‘u16’
does break strict-aliasing rules
include/rte_ip.h:157: note: initialized from here
...
The root cause is that, compile enable strict aliasing by default,
while in function rte_raw_cksum() try to convert 'const char *'
to 'const uint16_t *'.
This workaround is to solve the compile issue of GCC strict-aliasing (two
different type pointers should not be point to the same memory address).
For GCC 4.4.7 it will definitely occurs if flags "-fstrict-aliasing"
and "-Wall" used.
Signed-off-by: Michael Qiu <michael.qiu@intel.com>
[Thomas: add workaround comment]
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
lib/librte_eal/linuxapp/eal/eal.c(461): error #2259: non-pointer
conversion from "long long" to "void *" may lose significant bits
RTE_PTR_ALIGN_CEIL((uintptr_t)addr, RTE_PGSIZE_16M);
The root cause is that "RTE_PGSIZE_16M" is defined as unsigned long long.
But in i686 platform "void *" is 32-bit.
It is safe to cast to size_t and make it works in both 32 & 64-bit
platform.
Signed-off-by: Michael Qiu <michael.qiu@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
lib/librte_eal/linuxapp/eal/eal_memory.c:324:4: error: comparison
is always false due to limited range of data type [-Werror=type-limits]
|| (hugepage_sz == RTE_PGSIZE_16G)) {
^
This was introuduced by commit b77b5639:
mem: add huge page sizes for IBM Power
The root cause is that size_t is 32-bit in i686 platform,
but RTE_PGSIZE_16M and RTE_PGSIZE_16G are always 64-bit.
Force hugepage_sz to always 64-bit to avoid this issue.
Signed-off-by: Michael Qiu <michael.qiu@intel.com>
Suggested-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Include rte_memory.h for lib files that use __rte_cache_aligned
attribute.
Consider the following code:
struct per_core_foo {
...
} __rte_cache_aligned;
struct global_foo {
struct per_core_foo foo[RTE_MAX_CORE];
};
If __rte_cache_aligned is not defined (rte_memory.h is not included),
the code compiles but the structure is not aligned... it defines the
structure and creates a global variable called __rte_cache_aligned.
And this can lead to really bad things if this code is in a .h that
is included by files that may or may not include rte_memory.h
Signed-off-by: Jia Yu <jyu@vmware.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Add two macros I40E_VFTA_IDX and I40E_VFTA_BIT for vlan filter search and set.
Add vlan_id check in vlan filter search and set function.
Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jing Chen <jing.d.chen@intel.com>
">> 5" rather than ">> 4"
vlan id is a 12 bit value.
VFTA is 128 x 32 bit array (128 double word array) which could store 2^12 vlan bits.
Each bit represents whether corresponding vlan tag is set in the VSI.
Use high 7 bits as the index for the double word array.
Signed-off-by: Huawei Xie <huawei.xie@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jing Chen <jing.d.chen@intel.com>
The template mbuf_initializer is hard coded with a buflen which
might have been set differently by the application at the time of
mbuf pool creation.
- move buf_len fields out of rearm_data marker.
- make ixgbe_recv_pkts_vec() not touch buf_len field at all
(as all other RX functions behave).
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Jean-Mickael Guerin <jean-mickael.guerin@6wind.com>
Add a compiler barrier to make sure all fields covered by
the marker rearm_data are assigned before the read.
Fixes: 0ff3324da2 ("ixgbe: rework vector pmd following mbuf changes")
Signed-off-by: Jean-Mickael Guerin <jean-mickael.guerin@6wind.com>
Acked-by: David Marchand <david.marchand@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
The variable notify_pa is only initialized inside one branch of
an if statement, triggering a compiler error with clang 3.3 on FreeBSD.
CC vnic/vnic_dev.o
lib/librte_pmd_enic/vnic/vnic_dev.c:777:6: fatal error: variable 'notify_pa'
is used uninitialized whenever 'if' condition is false [-Wsometimes-uninitialized]
if (!vnic_dev_in_reset(vdev)) {
Fix this issue by adding "= 0" to the variable definition.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
This patch fixes the following compiler error raised by clang 3.3
on FreeBSD 10:
CC enic_clsf.o
lib/librte_pmd_enic/enic_clsf.c:99:25: fatal error: missing field 'u' initializer [-Wmissing-field-initializers]
struct filter fltr = {0};
It fixes it by changing the initializer to set a named field to zero,
thereby automatically setting the rest of the unnamed fields also to
zero.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
This allows the PMD to compile with kernels that don't support the
options in question. The "#if defined(...)" lines are a bit ugly,
but I don't know of any better way to accomplish the task.
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
When a lookup was done on a table_array structure with an incomplete
bitmask, the results was always zero hits. This was because the
pkts_mask value was cleared as we process each entry, and the result
was assigned at the end of the loop, when pkts_mask was zero.
Changing the assignment to occur at the start, before the pkts_mask
gets cleared, fixes this issue.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
In order not to affect the FVL's performance by default setting, this
patch moves the flow director initialization from i40e_pf_setup to
i40e_dev_configure according to the mode in fdir configure info.
Then the resources used for flow director will be only setup if it is enabled.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Replace the inner_l2_len and the inner_l3_len field with the
outer_l2_len and outer_l3_len field, and rework csum forward engine
and i40e PMD due to these changes.
Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Replace PKT_TX_VXLAN_CKSUM with PKT_TX_UDP_TUNNEL_PKT in order to indicate
a packet is an UDP tunneling packet, and introduce 3 TX offload flags for
outer IP TX checksum, which are PKT_TX_OUTER_IP_CKSUM, PKT_TX_OUTER_IPV4
and PKT_TX_OUTER_IPV6 respectively.
Rework csum forward engine and i40e PMD due to these changes.
Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
The reason of redefining the PKT_TX_IPV4 and the PKT_TX_IPV6 is listed below,
It will avoid to send a packet with a bad info:
- we receive a Ether/IP6/IP4/L4/data packet
- the driver sets PKT_RX_IPV6_HDR
- the stack decapsulates IP6
- the stack sends the packet, it has the PKT_TX_IPV6 flag but it's an IPv4 packet.
Signed-off-by: Jijiang Liu <jijiang.liu@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
There is no standard to check endianness.
So we need to try different checks.
Previous trials were done in testpmd (see commits
51f694dd40 and 64741f237c) without full success.
This one is not guaranteed to work everywhere so it could
evolve when exceptions are found.
If endianness is not detected, there is a fallback on x86
to little endian. It could be forced before doing detection
but it would add some arch-dependent code in the generic header.
The option CONFIG_RTE_ARCH_BIG_ENDIAN introduced for IBM Power only
(commit a982ec81d8) can be removed. A compile-time check is better.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: Michael Qiu <michael.qiu@intel.com>
When using test-pmd with flow director in FreeBSD, the application will
segfault/Bus error while parsing the command-line. This is due to how
each commands result structure is represented during parsing, where the offsets
for each tokens value is stored in a character array(char result_buf[BUFSIZ])
in cmdline_parse()(./lib/librte_cmdline/cmdline_parse.c).
The overflow occurs where BUFSIZ is less than the size of a commands result
structure, in this case "struct cmd_pkt_filter_result"
(app/test-pmd/cmdline.c) is 1088 bytes and BUFSIZ on FreeBSD is 1024 bytes as
opposed to 8192 bytes on Linux.
The problem can be reproduced by running test-pmd on FreeBSD:
./testpmd -c 0x3 -n 4 -- -i --portmask=0x3 --pkt-filter-mode=perfect
And adding a filter:
add_perfect_filter 0 udp src 192.168.0.0 1024 dst 192.168.0.0 1024 flexbytes
0x800 vlan 0 queue 0 soft 0x17
This patch removes the OS dependency on BUFSIZ and defines and uses a
library #define CMDLINE_PARSE_RESULT_BUFSIZE 8192
Added boundary checking to ensure this buffer size cannot overflow, with
an error message being produced.
Suggested-by: Olivier Matz <olivier.matz@6wind.com>
http://git.droids-corp.org/?p=libcmdline.git;a=commitdiff;h=b1d5b169352e57df3fc14c51ffad4b83f3e5613f
Signed-off-by: Alan Carew <alan.carew@intel.com>
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Tested-by: Bruce Richardson <bruce.richardson@intel.com>
Revert commit a0547e0a75 because it is an old version
of the patch and was applied by error.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
A lot of warnings were not seen because $(WERROR_FLAGS) was not set
in the Makefile. But they appear with toolchains that enforce more checks.
-Wno-deprecated seems useless.
-Wno-strict-aliasing is added to avoid false positives.
This patch cleans up unused variable, unused functions, wrong types,
static declarations, etc. A lot of functions have unused parameters;
it suggests that more clean-up could be needed.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Sujith Sankar <ssujith@cisco.com>
Because of different cache line size, the alignment of struct
rte_kni_mbuf in rte_kni_common.h doesn't work on IBM Power. This patch
changed from 64 to RTE_CACHE_LINE_SIZE micro to do the alignment.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
When using test-pmd with flow director in FreeBSD, the application will
segfault/Bus error while parsing the command-line. This is due to how
each commands result structure is represented during parsing, where the offsets
for each tokens value is stored in a character array(char result_buf[BUFSIZ])
in cmdline_parse()(./lib/librte_cmdline/cmdline_parse.c).
The overflow occurs where BUFSIZ is less than the size of a commands result
structure, in this case "struct cmd_pkt_filter_result"
(app/test-pmd/cmdline.c) is 1088 bytes and BUFSIZ on FreeBSD is 1024 bytes as
opposed to 8192 bytes on Linux.
This patch removes the OS dependency on BUFSIZ and defines and uses a
library #define CMDLINE_PARSE_RESULT_BUFSIZE 8192
The problem can be reproduced by running test-pmd on FreeBSD:
./testpmd -c 0x3 -n 4 -- -i --portmask=0x3 --pkt-filter-mode=perfect
And adding a filter:
add_perfect_filter 0 udp src 192.168.0.0 1024 dst 192.168.0.0 1024 flexbytes
0x800 vlan 0 queue 0 soft 0x17
Signed-off-by: Alan Carew <alan.carew@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
With this patch, KNI interface (e.g. vEth0) is created in the
network namespace where the DPDK application is running.
Otherwise, all interfaces are created in the default namespace
in the host.
put_net() is required, since get_net_ns_by_pid() increments
the reference counter of the network namespace with get_net().
Signed-off-by: Takayuki Usui <takayuki@midokura.com>
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
The compile error will occur as below when set 'RTE_LIBRTE_I40E_16BYTE_RX_DESC=y'.
'fd_id' should be used to replace 'fd', as 'fd' is not defined in that structure
at all. In addition, local variable of 'flexbl' and 'flexbh' must be used only if
32 bytes RX descriptor is selected.
error logs:
lib/librte_pmd_i40e/i40e_rxtx.c: In function i40e_rxd_build_fdir:
lib/librte_pmd_i40e/i40e_rxtx.c:431:28: error: volatile union <anonymous> has no member named fd
lib/librte_pmd_i40e/i40e_rxtx.c:427:19: error: unused variable flexbl [-Werror=unused-variable]
lib/librte_pmd_i40e/i40e_rxtx.c:427:11: error: unused variable flexbh [-Werror=unused-variable]
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Jingjing Wu <jingjing.wu@intel.com>
This patch fixes a maybe-uninitialized warning when compiling DPDK with
GCC 4.9 + Link Time Optimization.
Signed-off-by: Dennis Marinus <dmarinus@amazon.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Since commit aae1047905 ("use the right debug macro"),
DEBUGOUT was replaced by PMD_DRV_LOG which requires at least
2 arguments. But the level argument was missing.
Commit 7a10de5e27 fixed the logs but not the macros FUNC_PTR_OR_*
which are not preprocessed if RTE_LIBRTE_IXGBE_DEBUG_DRIVER is disabled.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: David Marchand <david.marchand@6wind.com>
This patch fixes the warnings and error reported by clang compiler on Linux.
Reported-by: Bruce Richardson <bruce.richardson@intel.com>
Signed-off-by: Sujith Sankar <ssujith@cisco.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Commit 1224decaa4 ("support TCP segmentation offload")
changed the way the bitfields are assigned in ixgbe, example:
tx_offload_mask.l2_len = ~0;
This result in a compilation error with clang:
error: implicit truncation from 'int' to bitfield
changes value from -1 to 127 [-Werror,-Wbitfield-constant-conversion]
Replacing the '=' with a '|=' fixes the issue.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
ENIC PMD was giving compilation errors on ppc_64-power8-linuxapp-gcc because
of types such as u_int32_t. This patch replaces all those with uint32_t and
similar ones.
Reported-by: David Marchand <david.marchand@6wind.com>
Signed-off-by: Sujith Sankar <ssujith@cisco.com>
Acked-by: David Marchand <david.marchand@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
GCC 4.3 complains that slow_pkts array in bond_ethdev_tx_burst_8023ad
may be used uninitialized, so it has been initialized to NULL.
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
The statistics that is reported through the rx_nombuf fields in struct
rte_eth_stats was not set when the vector PMD was used. The statistics
should report the number of mbufs that could _not_ be allocated during
rearm of the RX queue. The non-vector PMD reports it correctly. The
use of either vector PMD or non-vector PMD depends on runtime
configuration. Hence it is possible that a change in configuration
would disable this statistics. To prevent this from happening, the
statistics should be reported by both implementations.
Signed-off-by: Balazs Nemeth <balazs.nemeth@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Before the fix, bond device's offload capabilities are unset. This fix
takes the minimum common set of slave devices' capabilities as bond
device's capabilities. For simplicity, we ensure all slave devices
to have a capability before bond device can claim this capability,
even if some slave devices are unused (i.e. linked down, standby).
Signed-off-by: Jia Yu <jyu@vmware.com>
Signed-off-by: Declan Doherty <declan.doherty@intel.com>
Add support for mode 5 (Transmit load balancing) into pmd driver
This patch add support for Adaptive transmit load balancing (mode 5) to the
librte_pmd_bond library. This mode provides an adaptive transmit load
balancing. It dynamically changes the transmitting slave, according to the
computed load.
Further details are described here:
https://www.kernel.org/doc/Documentation/networking/bonding.txt
In implementation callback is used for sorting slave order - providing
statistics for burst function about slave bandwith usage and sort
interfaces due to usage.
Difference in this implementation vs Linux implementation:
- We Are trying send all pkts – If one interface hasn’t send packets we are
trying to send rest of packets by other slaves sorted previously by callback
function.
Some implementation details:
- Every 100ms is taken obytes statistics from every slave.
- Every 10 ms the slaves in table are sorted and updated by callback -
bandwidth and successfully transmitted bytes from previous iteration which
happens every 100 ms
- There is callback function which updates this statistics for transparency and
for rather intensive computation involved in this mode.
Test report: http://dpdk.org/ml/archives/dev/2014-November/008729.html
Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com>
Tested-by: SunX Jiajia <sunx.jiajia@intel.com>
Acked-by: Declan Doherty <declan.doherty@intel.com>
This patch set add support for dynamic link aggregation (mode 4) to the
librte_pmd_bond library. This mode provides auto negotiation/configuration
of peers and well as link status changes monitoring using out of band
LACP (link aggregation control protocol) messages. For further details of
LACP specification see the IEEE 802.3ad/802.1AX standards. It is also
described here
https://www.kernel.org/doc/Documentation/networking/bonding.txt.
In this implementation we have an array of mode 4 settings for each slave.
There is also assumption that for every port is one aggregator (it might
be unused if better is found).
Difference in this implementation vs Linux implementation:
- this implementation it is not directly based on state machines but current
state is calculated from actor and partner states (and other things too).
Some implementation details:
- during rx burst every packet Is checked if this is LACP or marker packet.
If it is LACP frame it is passed to mode 4 logic using slaves rx ring and
removed from rx buffer before it is returned
- in tx burst, packets from mode 4 (if any) are injected into each slave.
- there is a timer running in background to process/produce mode 4
frames form rx/to tx functions.
Some requirements for this mode:
- for LACP mode to work rx and tx burst functions must be invoked
at least in 100ms intervals
- provided buffer to rx burst should be at least 2x slave count size. This is
not needed but might increase performance especially during initial
handshake.
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Acked-by: Declan Doherty <declan.doherty@intel.com>
Inclusion of vfio.h was giving compilation errors if kernel version is less
than 3.6.0 and if RTE_EAL_VFIO was in config.
Removed inclusion of vfio.h and replaced RTE_EAL_VFIO with VFIO_PRESENT.
Reported-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Signed-off-by: Sujith Sankar <ssujith@cisco.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
It was impossible to include netinet/in.h and rte_ip.h
because the IP protocols were redefined.
It is removed because useless.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Ivan Boule <ivan.boule@6wind.com>
Check the FILE *f and rte_mempool *mp pointers for NULL.
Signed-off-by: Keith Wiles <keith.wiles@windriver.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
CACHE_LINE_SIZE is a macro defined in machine/param.h in FreeBSD and
conflicts with DPDK macro version.
Adding RTE_ prefix to avoid conflicts.
CACHE_LINE_MASK and CACHE_LINE_ROUNDUP are also prefixed.
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
[Thomas: updated on HEAD, including PPC]
The HPET support in the BSD EAL was copied directly from the Linux version,
but did not actually work on FreeBSD. We replace this old code with a simple
compiler message that informs the user that we don't support HPET on BSD if
they enable such support in the build-time configuration file.
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Signed-off-by: David Marchand <david.marchand@6wind.com>
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
RTE_ARCH_X86_64 can not be used as a way to determine if we are building for
64bits cpus. Instead, RTE_ARCH_64 should be used.
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Handle the RTE_ETH_FILTER_ADD and RTE_ETH_FILTER_DELETE operations
on ethertype filter.
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Jijiang Liu <jijiang.liu@intel.com>
A new structure of ethertype filter is defined in rte_eth_ctrl.h
for filter_ctrl api
Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
Acked-by: Jijiang Liu <jijiang.liu@intel.com>
Signed-off-by: Sujith Sankar <ssujith@cisco.com>
Acked-by: David Marchand <david.marchand@6wind.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
[Thomas: enable for BSD - not tested]
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
VNIC common code is partially shared with ENIC kernel mode driver.
Signed-off-by: Sujith Sankar <ssujith@cisco.com>
Acked-by: David Marchand <david.marchand@6wind.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
The mmap of hugepage files on IBM Power starts from high address to low
address. This is different from x86. This patch modified the memory
segment detection code to get the correct memory segment layout on Power
architecture. This patch also added a commond ARCH_PPC_64 definition for
64 bit systems.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@6wind.com>
IBM Power architecture has different huge page sizes (16MB, 16GB) than
x86.This patch defines RTE_PGSIZE_16M and RTE_PGSIZE_16G in the
rte_page_sizes enum variable and adds huge page size support of DPDK
for IBM Power architecture.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@6wind.com>
iopl() call is mostly for the i386 architecture. In Power and other
architecture, it doesn't exist. This patch modified rte_eal_iopl_init()
and make it return -1 for Power and other architecture. Thus
rte_config.flags will not contain EAL_FLG_HIGH_IOPL flag for other
architecture.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@6wind.com>
IBM Power processor doesn't have CPU flag hardware registers. This patch
uses aux vector software register to get CPU flags and add CPU flag
checking support for IBM Power architecture.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@6wind.com>
The SSE based memory copy in DPDK only support x86. This patch adds
altivec based memory copy functions for IBM Power architecture. This
patch includes altivec.h which requires GCC version>= 4.8.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@6wind.com>
This patch adds spinlock operations for IBM Power architecture.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@6wind.com>
This patch add architecture specific prefetch operations for IBM Power
architecture.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@6wind.com>
IBM Power architecture doesn't have TSC register to get CPU cycles. This
patch implements the time base register read instead of TSC register of
x86 on IBM Power architecture.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@6wind.com>
This patch adds architecture specific byte order operations for IBM Power
architecture. Power architecture support both big endian and little
endian. This patch also adds a RTE_ARCH_BIG_ENDIAN micro.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@6wind.com>
This patch adds architecture specific atomic operation file for IBM
Power architecture CPU.
Signed-off-by: Chao Zhu <chaozhu@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@6wind.com>
Implement TSO (TCP segmentation offload) in ixgbe driver. The driver is
now able to use PKT_TX_TCP_SEG mbuf flag and mbuf hardware offload infos
(l2_len, l3_len, l4_len, tso_segsz) to configure the hardware support of
TCP segmentation.
In ixgbe, when doing TSO, the IP length must not be included in the TCP
pseudo header checksum. A new function ixgbe_fix_tcp_phdr_cksum() is
used to fix the pseudo header checksum of the packet before giving it to
the hardware.
In the patch, the tx_desc_cksum_flags_to_olinfo() and
tx_desc_ol_flags_to_cmdtype() functions have been reworked to make them
clearer. This should not impact performance as gcc (version 4.8 in my
case) is smart enough to convert the tests into a code that does not
contain any branch instruction.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Some of the NICs supported by DPDK have a possibility to accelerate TCP
traffic by using segmentation offload. The application prepares a packet
with valid TCP header with size up to 64K and deleguates the
segmentation to the NIC.
Implement the generic part of TCP segmentation offload in rte_mbuf. It
introduces 2 new fields in rte_mbuf: l4_len (length of L4 header in bytes)
and tso_segsz (MSS of packets).
To delegate the TCP segmentation to the hardware, the user has to:
- set the PKT_TX_TCP_SEG flag in mbuf->ol_flags (this flag implies
PKT_TX_TCP_CKSUM)
- set the flag PKT_TX_IPV4 or PKT_TX_IPV6
- set PKT_TX_IP_CKSUM if it's IPv4, and set the IP checksum to 0 in
the packet
- fill the mbuf offload information: l2_len, l3_len, l4_len, tso_segsz
- calculate the pseudo header checksum without taking ip_len in account,
and set it in the TCP header, for instance by using
rte_ipv4_phdr_cksum(ip_hdr, ol_flags)
The API is inspired from ixgbe hardware (the next commit adds the
support for ixgbe), but it seems generic enough to be used for other
hw/drivers in the future.
This commit also reworks the way l2_len and l3_len are used in igb
and ixgbe drivers as the l2_l3_len is not available anymore in mbuf.
Signed-off-by: Mirek Walukiewicz <miroslaw.walukiewicz@intel.com>
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Introduce new functions to calculate checksums. These new functions
are derivated from the ones provided csumonly.c but slightly reworked.
There is still some room for future optimization of these functions
(maybe SSE/AVX, ...).
This API will be modified in tbe next commits by the introduction of
TSO that requires a different pseudo header checksum to be set in the
packet.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
In test-pmd (rxonly.c), the code is able to dump the list of ol_flags.
The issue is that the list of flags in the application has to be
synchronized with the flags defined in rte_mbuf.h.
This patch introduces 2 new functions rte_get_rx_ol_flag_name()
and rte_get_tx_ol_flag_name() that returns the name of a flag from
its mask. It also fixes rxonly.c to use this new functions and to
display the proper flags.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
This definition is specific to Intel PMD drivers and its definition
"indicate what bits required for building TX context" shows that it
should not be in the generic rte_mbuf.h but in the PMD driver.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Describe how to use hardware checksum API.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
The tx mbuf flags are now ordered from the lowest value to the
the highest. Add comments to explain where to add new flags.
By the way, move the PKT_TX_VXLAN_CKSUM at the right place.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Since commit 4332beee9 "mbuf: expand ol_flags field to 64-bits", the
packet flags are now 64 bits wide. Some occurences were forgotten in
the ixgbe driver.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
According to Intel® 82599 10 GbE Controller Datasheet (Table 7-38), both
L2 and L3 lengths are needed to offload the IP checksum.
Note that the e1000 driver does not need to be patched as it already
contains the fix.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
librte_power now contains both rte_power_acpi_cpufreq and rte_power_kvm_vm
implementations.
Signed-off-by: Alan Carew <alan.carew@intel.com>
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Provides a command packet format for host and guest.
Signed-off-by: Alan Carew <alan.carew@intel.com>
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Moved the current librte_power implementation to rte_power_acpi_cpufreq, with
renaming of functions only.
Added rte_power_kvm_vm implementation to support Power Management from a VM.
librte_power now hides the implementation based on the environment used.
A new call rte_power_set_env() can explicidly set the environment, if not
called then auto-detection takes place.
rte_power_kvm_vm is subset of the librte_power APIs, the following is supported:
rte_power_init(unsigned lcore_id)
rte_power_exit(unsigned lcore_id)
rte_power_freq_up(unsigned lcore_id)
rte_power_freq_down(unsigned lcore_id)
rte_power_freq_min(unsigned lcore_id)
rte_power_freq_max(unsigned lcore_id)
The other unsupported APIs return -ENOTSUP
Signed-off-by: Alan Carew <alan.carew@intel.com>
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Allows for the opening of Virtio-Serial devices on a VM, where a DPDK
application can send packets to the host based monitor. The packet formatted is
specified in channel_commands.h
Each device appears as a serial device in path
/dev/virtio-ports/virtio.serial.port.<agent_type>.<lcore_num> where each lcore
in a DPDK application has exclusive to a device/channel.
Each channel is opened in non-blocking mode, after a successful open a test
packet is send to the host to ensure the host side is monitoring.
Signed-off-by: Alan Carew <alan.carew@intel.com>
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Multi-process DPDK application must mmap hugepages and PCI resources
into the same virtual address space. By default the virtual addresses
are chosen by the primary process automatically when calling the mmap.
But sometimes the chosen virtual addresses aren't usable in secondary
process - for example, secondary process is linked with more libraries
than primary process, and the library occupies the same address space
that the primary process has requested for PCI mappings.
This patch makes EAL try and map PCI BARs right after the hugepages
(instead of location chosen by mmap) in virtual memory, so that PCI BARs
have less chance of ending up in random places in virtual memory.
Signed-off-by: Liang Xu <liang.xu@cinfotech.cn>
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Enable users to specify the lcore id that is used as master lcore.
Signed-off-by: Simon Kuenzer <simon.kuenzer@neclab.eu>
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
EAL -c option allows the user to enable any lcore in the system.
Often times, the user app wants to know 1st enabled core, 2nd
enabled core, etc, rather than phyical core ID (rte_lcore_id().)
The new API rte_lcore_index() will return an index from enabled lcores
starting from zero.
Signed-off-by: Patrick Lu <patrick.lu@intel.com>
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
In current version, used cores can only be specified using a bitmask.
It will now be possible to specify cores in 2 different ways:
- Using a bitmask (-c [0x]nnn): bitmask must be in hex format
- Using a list in following format: -l <c1>[-c2][,c3[-c4],...]
The letter -l can stand for lcore or list.
-l 0-7,16-23,31 being equivalent to -c 0x80FF00FF
Signed-off-by: Didier Pallard <didier.pallard@6wind.com>
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Some adjustments are done after options parsing and are common
to Linux and BSD.
Remove process_type adjustment in rte_config_init() because
it is already done in eal_parse_args().
eal_proc_type_detect() is kept duplicated because it open a
file descriptor which is used later in each eal.c.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
No need to have duplicated check for common options.
Some flags are set for options -c and -m in order to simplify the
checks.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Now that internal config structure is common to Linux and BSD,
we can have a common function to initialize it.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Some guards are missing or have a wrong name.
Others have LINUXAPP in their name but are now common.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
