Since DPDK 1.4, if RTE_EAL_UNBIND_PORTS is disabled, igb_uio mapping is
done for all devices (commit eee16c964c), breaking some non-Intel drivers.
But pci_uio_map_resource() should only be called for Intel devices
(using igb_uio kernel module).
The flag RTE_PCI_DRV_NEED_IGB_UIO is set for all those devices, even when
RTE_EAL_UNBIND_PORTS is disabled (fixes commit a22f5ce8fc).
Signed-off-by: David Marchand <david.marchand@6wind.com>
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Damien Millescamps <damien.millescamps@6wind.com>
Add a file app/test/test_kvargs.c that checks the rte_kvargs API.
The test passes:
RTE>>kvargs
== test valid case ==
== test invalid case ==
PMD: Error parsing device, invalid key <wrong-key>
Test OK
I also tested that rte_eth_pcap runs with the following arguments:
./app/testpmd -c 0x15 -n 3 --proc-type=primary --huge-dir=/mnt/huge \
--use-device="eth_pcap0;iface=ixgbe0" \
-- -i --port-topology=chained
./app/testpmd -c 0x15 -n 3 --proc-type=primary --huge-dir=/mnt/huge \
--use-device="eth_pcap0;rx_iface=ixgbe0;rx_iface=ixgbe1;tx_iface=ixgbe0" \
-- -i --port-topology=chained
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Adding or subtracting a value to a pointer makes a new pointer
of unknown type.
So typeof() is replaced by (void*) in RTE_PTR_ADD() and RTE_PTR_SUB().
But RTE_PTR_ALIGN_* macros have in their explicit API to return a pointer
of the same type. Since RTE_PTR_ALIGN_CEIL is based on RTE_PTR_ADD, a
typeof() is added to keep the original behaviour.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Hardware does not support RSS in SRIOV setups.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Signed-off-by: David Marchand <david.marchand@6wind.com>
The mq_mode was not set when rxq is > 0; it's defaulted to ETH_MQ_RX_NONE.
As a result, RSS remains inactive. The fix is to set mq_mode to ETH_MQ_RX_RSS
when hf is non-zero.
This bug was introduced by commit 243db2ddee
igb/ixgbe: ETH_MQ_RX_NONE should disable RSS
Signed-off-by: Daniel Kan <dan@nyansa.com>
Acked-by: Maxime Leroy <maxime.leroy@6wind.com>
This adds a new forwarding mode to testpmd, whereby it will forward
packets, but on a transmit failure, it will wait a small amount of time
before retrying the transmit again. This mode can be useful in some
cases to set a zero-packet loss rate at the cost of overall throughput,
especially in cases where TX ring size/buffering is constrained, such as
with virtio.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
* Change behaviour to spread the memory specified across all sockets
rather than duplicating the memory for each one. This way the memory
size specified is the total memory needed by the test.
* Minor reordering of test cases, to ensure that ring autotest is the
final test run.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
It's no longer possible to run all autotests in
a single go and have them pass, so remove that option.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Minor chanes including replacing local DIM macro with more standard
RTE_DIM macro from rte_common.h.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
don't compile the timer autotests unless timer library is turned on in
the compile-time configuration.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Minor fixes/updates to the autotest app, including:
* replace hard-coded values with #defines in rte_power.h
* replace local DIM() macro with common RTE_DIM
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Core support for using the Intel DPDK with Xen Dom0 - including EAL
changes and mempool changes. These changes encompass how memory mapping
is done, including support for initializing a memory pool inside an
already-allocated block of memory.
KNI sample app updated to use KNI close function when used with Xen.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
These library changes provide a new Intel DPDK feature for communicating
with virtual machines using QEMU's IVSHMEM mechanism.
The feature works by providing a command line for QEMU to map several hugepages
into a single IVSHMEM device. For the guest to know what is inside any given IVSHMEM
device (and to distinguish between Intel(R) DPDK and non-Intel(R) DPDK IVSHMEM
devices), a metadata file is also mapped into the IVSHMEM segment. No work needs to
be done by the guest application to map IVSHMEM devices into memory; they are
automatically recognized by the Intel(R) DPDK Environment Abstraction Layer (EAL).
Changes in this patch:
* Changes to EAL to allow mapping of all hugepages in a memseg into a single file
* Changes to EAL to allow ivshmem devices to be transparently mapped in
the process running on the guest.
* New ivshmem library to create and manage metadata exported to guest VM's
* New ivshmem compilation targets
* Mempool and ring changes to allow export of structures to a VM and allow
a VM to attach to those structures.
* New autotests to unit tests this functionality.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
For certain functionality, e.g. Xen Dom0 support, it is required that
we can guarantee that memzones for descriptor rings won't cross 2M
boundaries. So add new memzone reserve function where we can pass in a
boundary condition parameter.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Changes to allow compilation and use on FreeBSD. Includes:
* contigmem and nic_uio driver for FreeBSD
* new EAL instance
* new "bsdapp" compilation target
* various compilation fixes due to differences between linux and freebsd
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
This application is built if LIBRTE_CMDLINE is enabled.
But there was no enabled source file if APP_TEST is disabled.
Let's consider that CONFIG_RTE_APP_TEST apply only on app/test.
Signed-off-by: Intel
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
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
Fix warnings of type "Value stored to 'xxx' is never read".
Signed-off-by: Zijie Pan <zijie.pan@6wind.com>
Acked-by: Ivan Boule <ivan.boule@6wind.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
When a memory address must be read, for instance a [mapped] PCI register,
the read value is assigned to a local variable that is not used after,
as for instance:
x = *((uint8_t *) mem_addr);
Such instructions do not compile with gcc 4.6.
The fix consists in adding the "volatile" attribute to the accessed data type
and to not assign the read value:
*((volatile uint8_t *) mem_addr);
Signed-off-by: Ivan Boule <ivan.boule@6wind.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Build mbuf and lists of mbufs in a way compliant with the checks performed
by the function __rte_mbuf_sanity_check() when CONFIG_RTE_LIBRTE_MBUF_DEBUG=y
Signed-off-by: Ivan Boule <ivan.boule@6wind.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
A chained topology must always have an increment of 1.
Here, it was 2 if ports number is even.
Signed-off-by: Damien Millescamps <damien.millescamps@6wind.com>
Acked-by: Ivan Boule <ivan.boule@6wind.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
test-pmd txonly leaks mbuf from the pool.
The function tx_mbuf_alloc() does not change the refcnt
and the refcnt is 0 when it is first allocated.
However, rte_pktmbuf_free_seg called by the driver's xmit code decrements
reference count to -1. So mbuf never goes back to the pool.
As a result, txonly can't send packets after it exhausts the mempool.
The function tx_mbuf_alloc() was getting mbuf directly from mempool and so
was bypassing mbuf API.
The dedicated function is rte_pktmbuf_alloc() but it is much slower because
it does unnecessary initializations in rte_pktmbuf_reset().
By using the internal API __rte_mbuf_raw_alloc(), refcnt is correctly handled
without adding too much overload.
Signed-off-by: Dongsu Han <dongsuh@cs.cmu.edu>
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
