Intel [Wed, 18 Sep 2013 10:00:00 +0000 (12:00 +0200)]
timer: use a skip list
The skip list algorithm allows to improve the scalability.
Signed-off-by: Intel
Intel [Wed, 18 Sep 2013 10:00:00 +0000 (12:00 +0200)]
lpm: rework rules storage
Signed-off-by: Intel
Intel [Wed, 18 Sep 2013 10:00:00 +0000 (12:00 +0200)]
sched: minor changes
Do not define grinder_credits_check() if it is not used.
Signed-off-by: Intel
Intel [Wed, 18 Sep 2013 10:00:00 +0000 (12:00 +0200)]
eal: use pause only with SSE2
The pause instruction is part of SSE2 extensions.
Note that some compilers define _mm_pause as "rep; nop" instead of "pause".
For compatible processors, they are equivalent.
http://www.intel.com/Assets/PDF/manual/325383.pdf:
"
When executing a spin-wait loop, a Pentium 4 or Intel Xeon processor suffers
a severe performance penalty when exiting the loop because it detects a
possible memory order violation.
The PAUSE instruction provides a hint to the processor that the code sequence
is a spin-wait loop. The processor uses this hint to avoid the memory order
violation in most situations, which greatly improves processor performance.
"
Signed-off-by: Intel
Intel [Wed, 18 Sep 2013 10:00:00 +0000 (12:00 +0200)]
eal: allow to whitelist devices
The new option --use-device is a PCI whitelist.
It is the opposite to -b option.
Signed-off-by: Intel
Intel [Wed, 18 Sep 2013 10:00:00 +0000 (12:00 +0200)]
eal: allow to blacklist address without domain prefix
These 2 formats are now accepted:
domain:bus:device.function
bus:device.function
Signed-off-by: Intel
Intel [Wed, 18 Sep 2013 10:00:00 +0000 (12:00 +0200)]
eal: rework CPU mask parsing
The CPU mask was limited to "unsigned long long".
The limit was removed and parsing/init is less loosy.
Signed-off-by: Intel
Intel [Wed, 18 Sep 2013 10:00:00 +0000 (12:00 +0200)]
eal: minor changes
Signed-off-by: Intel
Intel [Wed, 18 Sep 2013 10:00:00 +0000 (12:00 +0200)]
mk: combined library
Allow to merge all libraries (shared or static) into one.
Signed-off-by: Intel
Intel [Wed, 18 Sep 2013 10:00:00 +0000 (12:00 +0200)]
mk: shared libraries
Allow to build shared libraries (.so) instead of static ones (.a).
Signed-off-by: Intel
Intel [Wed, 18 Sep 2013 10:00:00 +0000 (12:00 +0200)]
mk: link with libm
Signed-off-by: Intel
Intel [Wed, 18 Sep 2013 10:00:00 +0000 (12:00 +0200)]
doc: whitespace changes in licenses
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
update version to 1.4.1
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
examples/ip_reassembly: various updates
- postpone calls to rte_pktmbuf_free() when a mbuf is not used anymore
- add some tx statistics
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
examples/dpdk_qat: rework port discovery
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
examples/qos_sched: minor changes
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
examples/qos_sched: add --msz for mempool size
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
examples/l3fwd-power: use DD bit rather than RX queue count
Prefer use of rte_eth_rx_descriptor_done to rte_eth_rx_queue_count.
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
ixgbe: fix DCB setup
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
ixgbe: check DD bit for specific RX descriptor
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
igb: check DD bit of specific RX descriptor
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
ethdev: check DD bit of specific RX descriptor
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
ixgbe: RX queue count is not implemented for VF
It was introduced by mistake in version 1.4.0.
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
ixgbe: use DD bit to count RX available descriptors
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
igb: use DD bit to count RX available descriptors
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
ethdev: fix doxygen comment for rte_eth_rx_queue_count
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
pci: check driver probe return code
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
mem: rework huge page mapping for secondary process
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
sched: only support TC 3 oversubscription
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
sched: add mtu parameter
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
sched: bitmap is now dynamically allocated
Signed-off-by: Intel
Intel [Mon, 22 Jul 2013 22:00:00 +0000 (00:00 +0200)]
eal: use intrinsic function
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
update version to 1.4.0
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
examples/quota_watermark: initial import
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
examples/ip_reassembly: initial import
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
examples/l3fwd-vf: release resources on SIGINT
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
examples/l3fwd: various updates
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
examples/l2fwd: if no port configured, exit
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
examples/vmdq_dcb: display port mac address
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
examples/dpdk_qat: minor change
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
examples/dpdk_qat: fix cross compile check
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
examples: work only on defined lcores
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
examples: time fixes
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
examples: use global RTE_MAX_ETHPORTS
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
examples: numa updates
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
app/testpmd: update csumonly
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
app: fix build with GCC 4.8
GCC 4.8 was producing this error:
argument to ‘sizeof’ in ‘strncmp’ call is the same expression as the
second source; did you mean to provide an explicit length?
[-Werror=sizeof-pointer-memaccess]
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
app: only use one thread to run autotests
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
app: fix build when some libs are disabled
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
app: rework ring tests
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
app: add a test on mbuf alignment
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
app: add tests on --syslog, --no-shconf and --huge-dir flags
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
app: add some tests
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
app: remove pmac from Makefile
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
app: fix Makefile
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
kni: disable ETHTOOL_GRXRINGS
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
kni: fix build with kernel 3.8
- timecompare (used for hardware timestamping) has been removed.
(see Linux commit
65f8f9a1c1db831e5159e3e3e50912d1f214cd0c)
Simply disable HW_TIME_STAMP feature because it is not used by KNI.
- annotations __devinit and __devexit have been removed.
(see Linux commit
54b956b903607f8f8878754dd4352da6a54a1da2)
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
kni: fix multi-process support
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ixgbe: force phy power up/down
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
igb: force phy power up/down
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ixgbe: update VMDq/DCB support
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
igb: update VMDq/DCB support
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ethdev: update VMDq/DCB support
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ethdev: set VMDq pool when adding mac address
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ixgbe: add PF support
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
igb: add PF support
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ethdev: add PF support
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ixgbe: support dual vlan
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
igb: fix max RX packet size and support dual VLAN
When in 'normal' mode, hardware is designed to receive up to 1522 bytes.
When in 'jumbo' mode, RLPML register must be updated so that hardware accepts
a 802.1q vlan header. If dual vlan is enabled, then a second vlan header is
expected, so update RLPML register according to dual vlan support.
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ixgbe: retrieve RX available descriptors
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (02:00 +0200)]
igb: retrieve RX available descriptors
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (02:00 +0200)]
e1000: retrieve RX available descriptors
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ethdev: retrieve RX available descriptors
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ixgbe: RSS RETA configuration
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
igb: RSS RETA configuration
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ethdev: RSS RETA configuration
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ixgbe: allocate structures on numa_node
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ethdev: store numa_node per device
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ixgbe: allow unsupported SFP
Add RTE_LIBRTE_IXGBE_ALLOW_UNSUPPORTED_SFP build option
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ixgbe: force inlining
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ixgbe: log tx code path
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
e1000: move workaround for wthresh on 82576 at wrong place
The workaround is moved in e1000em code but 82576 is an igb NIC.
It breaks the fix of the commit
7e9e49feea.
Signed-off-by: Intel <intel.com>
Stephen Hemminger [Thu, 30 May 2013 17:12:35 +0000 (10:12 -0700)]
igb: workaround errata with wthresh on 82576
The 82576 has known issues which require the write threshold to be set to 1.
See:
http://download.intel.com/design/network/specupdt/82576_SPECUPDATE.pdf
If not then single packets will hang in transmit ring until more arrive.
Simple tests like ping will fail.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Reviewed-by: Vincent Jardin <vincent.jardin@6wind.com>
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ethdev: more doxygen comments about fdir
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
pci: store numa_node per device
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
pci: configure SR-IOV with max_vfs
Signed-off-by: Intel
Julien Courtat [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
pci: fix probing blacklisted device with RTE_PCI_DRV_MULTIPLE flag
When blacklisting network adapters that share a PCI bus address with multiple
ports such as Mellanox ConnectX-3 (librte_pmd_mlx4), RTE applications
enter infinite loops while probing PCI devices.
This commit removes this probe on subsequent instances for blacklisted devices
that have RTE_PCI_DRV_MULTIPLE set.
Signed-off-by: Julien Courtat <julien.courtat@6wind.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
pci: don't unbind resources on exit
unbind operations must be done before starting eal application
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
pci: rework interrupt handling
Signed-off-by: Intel <intel.com>
Stephen Hemminger [Thu, 30 May 2013 17:12:39 +0000 (19:12 +0200)]
pci: support multiple PCI regions per device
Need to change PCI code to support multiple I/O regions on a single device.
Some devices like VMXNET3 have multiple PCI memory regions, and some
have none.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
mbuf: force inlining
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
mem: force inlining
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
mem: remove RLIMIT_NOFILE update
Memory init won't try to increase its RLIMIT_NOFILE limit, this must be done
prior to starting it.
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
mem: find most little element from heap
Allocations now return the most little element from heap matching size
constraint instead of the first one available.
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
ring: optimisations
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
timer: add a rte_pause in a tight loop
Signed-off-by: Intel
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
timer: prefer TSC to HPET
Signed-off-by: Intel
Stephen Hemminger [Thu, 30 May 2013 17:12:36 +0000 (17:12 +0000)]
timer: optimize for empty case
In many application there are no timers queued, and the call to
rte_timer_managecan be optimized in that case avoid reading HPET and
lock overhead.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Reviewed-by: Vincent Jardin <vincent.jardin@6wind.com>
Damien Millescamps [Fri, 21 Sep 2012 11:36:40 +0000 (11:36 +0000)]
timer: option --vmware-tsc-map for VMware guest
The VMWare TSC mapping uses a hook to RDPMC to read the physical TSC
in the case of VMware ESXi.
Signed-off-by: Damien Millescamps <damien.millescamps@6wind.com>
Acked-by: Jean-Mickael Guerin <jmg@6wind.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Introduce new option --vmware-tsc-map, ignored if
CONFIG_RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT is not set.
Default is CONFIG_RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT=y.
if CONFIG_RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT is set:
rte_rdtsc() selects at runtime between Vmware mapping of
TSC or native TSC
else
rte_rdtsc() always uses native rdtsc.
When running DPDK on VMware guest, enable --vmware-tsc-map to
read the physical TSC.
Caution: ESXi should pass monitor_control.pseudo_perfctr = TRUE
othewise it results in general protection fault.
Signed-off-by: Jean-Mickael Guerin <jean-mickael.guerin@6wind.com>
Acked-by: Vincent Jardin <vincent.jardin@6wind.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Intel [Mon, 3 Jun 2013 00:00:00 +0000 (00:00 +0000)]
hash: jhash optimisation
Signed-off-by: Intel