Remove one limitation of rte_pktmbuf_attach(): "mbuf we're attaching to
must be direct".
Now, when we attach to an indirect mbuf:
- copy the all relevant fields (addr, len, offload, ...) as before
- get the pointer to the mbuf that embeds the data buffer (direct mbuf),
and increase the reference counter of this one.
When detaching the mbuf, we can retrieve this direct mbuf as the pointer
is determined from the buffer address.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Add a new priv_size field in mbuf structure that should
be initialized at mbuf pool creation. This field contains the
size of the application private data in mbufs.
Introduce new static inline functions rte_mbuf_from_indirect()
and rte_mbuf_to_baddr() to replace the existing macros, which
take the private size in account when attaching and detaching
mbufs.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Reviewed-by: Zoltan Kiss <zoltan.kiss@linaro.org> Acked-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
When it's possible, use the new helper to create the mbuf pools.
Most of the patch is trivial, except for the following files that
have some specifics (indirect mbufs):
- ip_fragmentation
- ip_pipeline
- ipv4_multicast
- vhost
Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
app/testpmd: use standard functions to initialize mbufs and mbuf pool
The rte_pktmbuf_pool_init() and rte_pktmbuf_init() functions now
support to have a non-hardcoded buffer length. We can remove the
specific functions used in testpmd and replace them by the standard
ones.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
examples: always initialize mbuf pool private area
The mbuf pool private area must always be populated in a mbuf pool.
The applications or drivers may expect that for a mbuf pool, the mbuf
pool private area (mbuf_data_room_size and mbuf_priv_size) are
properly filled.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Deduct the mbuf data room size from mempool->elt_size and priv_size,
instead of using an hardcoded value that is not related to the real
buffer size.
To use rte_pktmbuf_pool_init(), the user can either:
- give a NULL parameter to rte_pktmbuf_pool_init(): in this case, the
private size is assumed to be 0, and the room size is
mp->elt_size - sizeof(struct rte_mbuf).
- give the rte_pktmbuf_pool_private filled with appropriate
data_room_size and priv_size values.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
The statistics tables and null mac address should be immutable.
Fix up get_addr routines to accept const args.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
There is no reason to inline functions that are not in the critical
path.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
This patch helps enic migrate to the new flow-director API.
It takes care of the following.
1. The change in fdir_filter structure and stats structure
2. DPDK interface functions in enic_ethdev.c
3. ENIC driver functions that deal with the VIC adapter
Thomas Monjalon [Wed, 15 Apr 2015 17:09:28 +0000 (19:09 +0200)]
use simple zero initializers
To initialize a structure with zeros, one field was explicitly set
to avoid "missing initializer" bug with old GCC (e.g. 4.4).
This warning is now disabled (commit <insertlater>) for old versions of GCC,
so the workarounds may be removed.
These initializers should not be needed for static variables but they
are still used to workaround an ICC bug (see commit b2595c4aa92d).
There is one remaining exception where {0} initializer doesn't work cleanly,
even with recent GCC:
lib/librte_pmd_ixgbe/ixgbe_rxtx_vec.c:735:9:
error: missing braces around initializer [-Werror=missing-braces]
struct rte_mbuf mb_def = {0}; /* zeroed mbuf */
Tested with gcc-4.4.7 (CentOS), gcc-4.7.2 (Debian), gcc-4.9.2 (Arch),
clang-3.6.0 and icc-13.1.1.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com> Tested-by: Thomas Monjalon <thomas.monjalon@6wind.com> Tested-by: John McNamara <john.mcnamara@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Vlad Zolotarov <vladz@cloudius-systems.com>
Thomas Monjalon [Tue, 14 Apr 2015 09:23:36 +0000 (11:23 +0200)]
mk: fix build with gcc 4.4 and clang
With GCC 4.4.7 from CentOS 6.5, the following errors arise:
lib/librte_pmd_ixgbe/ixgbe_rxtx.c: In function 'ixgbe_dev_rx_queue_setup':
lib/librte_pmd_ixgbe/ixgbe_rxtx.c:2509: error: missing initializer
lib/librte_pmd_ixgbe/ixgbe_rxtx.c:2509: error: (near initialization for 'dev_info.driver_name')
lib/librte_pmd_ixgbe/ixgbe_rxtx.c: In function 'ixgbe_set_rsc':
lib/librte_pmd_ixgbe/ixgbe_rxtx.c:4072: error: missing initializer
lib/librte_pmd_ixgbe/ixgbe_rxtx.c:4072: error: (near initialization for 'dev_info.driver_name')
lib/librte_pmd_ixgbe/ixgbe_rxtx.c: In function 'ixgbe_recv_pkts_lro_single_alloc':
lib/librte_pmd_ixgbe/ixgbe_rxtx.c:1479: error: 'next_rsc_entry' may be used uninitialized in this function
lib/librte_pmd_ixgbe/ixgbe_rxtx.c:1480: error: 'next_rxe' may be used uninitialized in this function
The "missing initializer" warning is a GCC bug which seems fixed in 4.7.
The same warning is thrown by clang.
The "may be used uninitialized" warning is another GCC bug which seems fixed in 4.7.
Fixes: 8eecb3295aed ("ixgbe: add LRO support") Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Vlad Zolotarov <vladz@cloudius-systems.com>
Igor Ryzhov [Wed, 8 Apr 2015 16:31:07 +0000 (19:31 +0300)]
doc: fix vhost guide
Guide says that a configure parameter to choose between vhost cuse
and vhost user will be introduced in the future, but it’s already
added by commit 28a1ccca41bf.
Signed-off-by: Igor Ryzhov <iryzhov@nfware.com> Acked-by: Changchun Ouyang <changchun.ouyang@intel.com>
Vlad Zolotarov [Mon, 30 Mar 2015 19:21:11 +0000 (22:21 +0300)]
ixgbe: add LRO support
- Only x540 and 82599 devices support LRO.
- Add the appropriate HW configuration.
- Add RSC aware rx_pkt_burst() handlers:
- Implemented bulk allocation and non-bulk allocation versions.
- Add LRO-specific fields to rte_eth_rxmode, to rte_eth_dev_data
and to ixgbe_rx_queue.
- Use the appropriate handler when LRO is requested.
Signed-off-by: Vlad Zolotarov <vladz@cloudius-systems.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Vlad Zolotarov [Mon, 30 Mar 2015 19:21:10 +0000 (22:21 +0300)]
ixgbe: code refactoring
- ixgbe_rx_alloc_bufs():
- Reset the rte_mbuf fields only when requested.
- Take the RDT update out of the function.
- Add the stub when RTE_LIBRTE_IXGBE_RX_ALLOW_BULK_ALLOC is not defined.
- ixgbe_recv_scattered_pkts():
- Take the code that updates the fields of the cluster's HEAD buffer into
the inline function.
Signed-off-by: Vlad Zolotarov <vladz@cloudius-systems.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
The test PMD uses a special type of eth_dev_ops to test features.
Rather allocating this separately, just put in the private data area.
This allows for next change to make dev_ops const.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Neil Horman <nhorman@tuxdriver.com>
Thomas Monjalon [Thu, 9 Apr 2015 17:03:14 +0000 (19:03 +0200)]
scripts: test null forwarding
This script ease testing of basic initializations and Rx/Tx bursts.
It may help to check obvious regressions.
In order to run it on a standard development machine, it doesn't use
neither hugepages nor real interfaces.
The optional parameters are:
- build directory (default: build)
- coremask (default: 3 i.e. cores 0 and 1)
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Thomas Monjalon [Thu, 9 Apr 2015 17:23:45 +0000 (19:23 +0200)]
mk: remove uio suffix from virtio pmd
The virtio pmd is not restricted to uio anymore.
Fixes: da978dfdc43b ("virtio: use port IO to get PCI resource") Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com> Acked-by: Changchun Ouyang <changchun.ouyang@intel.com>
Thomas Monjalon [Thu, 2 Apr 2015 10:49:34 +0000 (12:49 +0200)]
pkg: update RPM
Enable vhost-user and build PDF doc.
Inkscape and TeXLive are required to convert .svg and .rst to .pdf.
Temporary sphinx files .* (.doctrees/ and .buildinfo) are cleaned.
Tested on Fedora 20.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Thomas Monjalon [Tue, 31 Mar 2015 12:45:06 +0000 (14:45 +0200)]
mbuf: clean old refcnt option
CONFIG_RTE_MBUF_SCATTER_GATHER was renamed into CONFIG_RTE_MBUF_REFCNT
by commit 62814bc2e923 and removed by commit 4769bc5a27cc.
Some traces remain because of delayed patches.
It can also be removed from doxygen config.
It is now poisoned in rte_mbuf.h to warn any misuse.
- libmlx4 and libibverbs dependencies distributed with Mellanox OFED are now
also available on DPDK.org to make installation easier.
- Document Mellanox OFED and firmware versions to use.
- Add links to Mellanox and its community websites.
- Add kernel modules parameters section.
Added notes for i40e firmware version. As base driver to support the
latest version of firmware (FVL3E) hasn't been integrated, currently
the validated version of firmware is 4.2.6.
API *rte_eth_link_get* expect to call a wait to complete link_update.
That's the difference between *rte_eth_link_get_nowait*.
The patch fixes the issue that i40e link_update ignores the wait_to_complete flag.
The issue impacts those applications calling rte_eth_link_get to get wrong intermediate link status.
Jingjing Wu [Thu, 12 Feb 2015 11:22:23 +0000 (19:22 +0800)]
i40e: fix out of bound read
Klocwork reports array 'src_offset' may use index 16.
In function i40e_srcoff_to_flx_pit, index j + 1 can reach I40E_FDIR_MAX_FLEX_LEN.
This patch fixes this issue to avoid array bound.
Test report: http://www.dpdk.org/ml/archives/dev/2015-March/016030.html
Fixes: d8b90c4eabe9 ("i40e: take flow director flexible payload configuration") Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com> Tested-by: Min Cao <min.cao@intel.com>
Pablo de Lara [Wed, 1 Apr 2015 12:10:11 +0000 (13:10 +0100)]
doc: add note for --txqflags in testpmd
Since txqflags is now set from the default rx/tx configuration,
depending on the PMD, it might not be 0.
Therefore, user has to overwrite it with --txqflags 0.
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com> Acked-by: Siobhan Butler <siobhan.a.butler@intel.com>
John McNamara [Wed, 1 Apr 2015 11:32:45 +0000 (12:32 +0100)]
mk: fix pdf doc rules order
Re-ordered the guide-pdf-% and guide-% rules to allow
the doc-guides-pdf target to build with make 3.81. It
builds unmodified with all later versions of make.
Fixes: ebf8050afd44 ("doc: add pdf output") Reported-by: Tetsuya Mukawa <mukawa@igel.co.jp> Signed-off-by: John McNamara <john.mcnamara@intel.com> Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Thomas Monjalon [Tue, 31 Mar 2015 22:00:37 +0000 (00:00 +0200)]
pci: rename passthrough driver to kernel driver
Kernel driver (kdrv) seems easier to understand than
passthrough driver (pt_driver). It's also more generic
as a PMD could run on top of any PCI kernel driver if
it would offer such support.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com> Acked-by: Michael Qiu <michael.qiu@intel.com> Acked-by: David Marchand <david.marchand@6wind.com>
Siobhan Butler [Thu, 12 Mar 2015 12:13:54 +0000 (12:13 +0000)]
doc: update release notes for new sample apps
Updated release notes release description notes:
- added new sample applications to list: Link Bonding,Skeleton, Callbacks, Jobstats
- updated copyright date to 2015
Bruce Richardson [Mon, 23 Mar 2015 16:19:50 +0000 (16:19 +0000)]
doc: add note on needing igb_uio for VF devices
Since the uio_pci_generic module requires that the device to which it is
being bound supports legacy interrupts, there can be problems using it
with VF devices. Add a note to the GSG doc to document this fact, and
provide information on loading igb_uio as a replacement.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Siobhan Butler <siobhan.a.butler@intel.com>
Olivier Matz [Thu, 19 Mar 2015 16:35:12 +0000 (17:35 +0100)]
doc: fix prefix in file references
There are some references to DPDK.xyz.mk files that do not exist in the
dpdk tree. This was probably the result of an automatic replacement, so
restore the proper file names which are rte.xyz.mk.