Add module parameter 'carrier='on|off' to set the default carrier state
for linux network interfaces created by the KNI module. The default
carrier state is 'off'.
For KNI interfaces which need to reflect the carrier state of
a physical Ethernet port controlled by the DPDK application, the
default carrier state should be left set to 'off'. The application
can set the carrier state of the KNI interface to reflect the state
of the physical Ethernet port using rte_kni_update_link().
For KNI interfaces which are purely virtual, the default carrier
state can be set to 'on'. This enables the KNI interface to be
used without having to explicity set the carrier state to 'on'
using rte_kni_update_link().
Signed-off-by: Dan Gora <dg@adax.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Adding memory barrier to make sure the values being synced
before updating fifo_write in kni_fifo_put and fifo_read in
kni_fifo_get.
Fixes: 3fc5ca2f63 ("kni: initial import")
Cc: stable@dpdk.org
Signed-off-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
A Kbuild is also included to allow users to use DKMS natively without
additional code.
Signed-off-by: Luca Boccassi <bluca@debian.org>
Tested-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
The build error observed with Linux kernel 4.19 when KNI ethtool
support enabled (CONFIG_RTE_KNI_KMOD_ETHTOOL=y)
.../build/build/kernel/linux/kni/kni_ethtool.c:193:3:
error: ‘struct ethtool_ops’ has no member named ‘get_settings’;
.get_settings = kni_get_settings,
^~~~~~~~~~~~
.../build/build/kernel/linux/kni/kni_ethtool.c:194:3:
error: ‘struct ethtool_ops’ has no member named ‘set_settings’;
.set_settings = kni_set_settings,
^~~~~~~~~~~~
With kernel 4.19 ethtool_ops `get_settings` & `set_settings` are
replaced with `get_link_ksettings` & `set_link_ksettings`
Commit 9b3004953503 ("ethtool: drop get_settings and set_settings callbacks")
This fix practically removes `get_settings` & `set_settings` support
for the kernel versions that have the new ethtool_ops without
implementing the new ones.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
When a device is hot-unplugged, pci_remove will be invoked unexpectedly
before pci_release, it will caused kernel hung issue which will throw the
error info of "Trying to free already-free IRQ XXX". And on the other hand,
if pci_remove before pci_release, the interrupt will not got chance to be
disabled. So this patch aim to fix this issue by adding pci_release call
in pci_remove, it will gurranty that all pci clean up will be done before
pci removal.
Signed-off-by: Jeff Guo <jia.guo@intel.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
This fixes the problem of reference count leak if
igbuio_pci_enable_interrupts fails.
Also, replace mutex and integer with a kernel atomic counter.
This is standard pattern for kernel devices.
Fixes: 19685d5aa7 ("igb_uio: allow multi-process access")
Cc: stable@dpdk.org
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Modify kni_net_ioctl() to return -EOPNOTSUPP for all ioctls instead
of 0.
This is necessary because the Wicked (and possibly other) network
interface managers will perform the SIOCGIWNAME ioctl to check if
the interface is a wireless interface. If the KNI module returns
success, Wicked will incorrectly interpret the interface as a wireless
interface.
Signed-off-by: Dan Gora <dg@adax.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Install in $kerneldir/../extra/dpdk. Usually $kerneldir should something
like: /lib/modules/$kver/build, so this directory will match the default
one used by legacy makefiles.
Fixes: a52f4574f7 ("igb_uio: build with meson")
Cc: stable@dpdk.org
Signed-off-by: Luca Boccassi <bluca@debian.org>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
This patch fixes compilation errors on Centos 7.5 when
CONFIG_RTE_KNI_KMOD_ETHTOOL is set to 'y'.
On RHEL75 ndo_change_mtu has changed to ndo_change_mtu_rh74.
See commit 37d477b686 ("kni: fix build on RHEL 7.5")
Signed-off-by: Drocula Lambda <quzeyao@gmail.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Caught on ubuntu-16.04 with hwe kernel for aarch64:
$ uname -a
Linux ubuntu1604arm64es 4.13.0-43-generic #48~16.04.1-Ubuntu SMP Thu May
17 13:08:01 UTC 2018 aarch64 aarch64 aarch64 GNU/Linux
== Build kernel/linux/igb_uio
CC [M] .../kernel/linux/igb_uio/igb_uio.o
In file included from .../kernel/linux/igb_uio/igb_uio.c:20:0:
.../igb_uio/compat.h: In function ‘igbuio_kernel_is_locked_down’:
.../igb_uio/compat.h:146:7:
error: "CONFIG_EFI_SECURE_BOOT_LOCK_DOWN" is not defined [-Werror=undef]
#elif CONFIG_EFI_SECURE_BOOT_LOCK_DOWN
^
cc1: all warnings being treated as errors
Fixes: d67014c3d3 ("igb_uio: fail and log if kernel lock down is enabled")
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Write combining (WC) increases NIC performance by making better
utilization of PCI bus, but cannot be use by all PMD.
To get internal_addr memory need to be mapped. But as memory could not be
mapped twice: with and without WC it should be skipped for WC. [1]
To do not spoil other drivers that potentially could use internal_addr,
parameter wc_activate adds possibility to skip it for those PMDs,
that do not use it.
[1] https://www.kernel.org/doc/ols/2008/ols2008v2-pages-135-144.pdf
section 5.3 and 5.4
Signed-off-by: Rafal Kozik <rk@semihalf.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
When EFI secure boot is enabled, it is possible to lock down kernel and
prevent accessing device BARs and this makes igb_uio unusable.
Lock down patches are not part of the vanilla kernel but they are
applied and used by some distros already [1].
It is not possible to fix this issue, but intention of this patch is to
detect and log if kernel lock down enabled and don't insert the module
for that case.
The challenge is since this feature enabled by distros, they have
different config options and APIs for it. This patch is done based on
Fedora and Ubuntu kernel source, may needs to add more distro specific
support.
[1]
kernel.ubuntu.com/git/ubuntu/ubuntu-artful.git/commit/?id=99f9ef18d5b6
And a few more patches too.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Luca Boccassi <bluca@debian.org>
Error observed when CONFIG_RTE_KNI_KMOD_ETHTOOL config option is
enabled.
build error:
In function ‘strncpy’,
inlined from ‘igb_get_drvinfo’ at
kernel/linux/kni/igb_ethtool.c:814:2:
include/linux/string.h:246:9: error: ‘__builtin_strncpy’ output
may be truncated copying 31 bytes from a string of length 42
[-Werror=stringop-truncation]
return __builtin_strncpy(p, q, size);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Fixed by using strlcpy instead of strncpy.
adapter->fw_version size kept same because of
c369819294 ("kni: fix build with gcc 7.1")
Also next line strncpy usage replaced with strlcpy while arround.
Fixes: c369819294 ("kni: fix build with gcc 7.1")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Fix compilation errors on Centos 7.4 when CONFIG_RTE_KNI_KMOD_ETHTOOL
is set to 'y'.
Centos 7.4 needs HAVE_NDO_BRIDGE_GETLINK_FILTER_MASK_VLAN_FILL:
igb_main.c: In function ‘igb_ndo_bridge_getlink’:
igb_main.c:2289:2: error: too few arguments to function
‘ndo_dflt_bridge_getlink’
return ndo_dflt_bridge_getlink(skb, pid, seq, dev, mode, 0, 0, nlflags);
^
Centos 7.4 needs HAVE_VF_VLAN_PROTO and needs to redefine
ndo_set_vf_vlan to .extended.ndo_set_vf_vlan:
igb_main.c:2318:2: error: unknown field ‘ndo_set_vf_vlan’ specified
in initializer
.ndo_set_vf_vlan = igb_ndo_set_vf_vlan,
^
Signed-off-by: Dan Gora <dg@adax.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
rx_q fifo can only be released by kernel thread. There may be
mbuf leaks in rx_q because kernel threads are randomly stopped.
When the kni is released and netdev is unregisterd, convert the
physical address mbufs in rx_q to the virtual address in free_q.
By the way, alloc_q can be processed together to speed up the
release rate in userspace.
In my test, it is improved from 300-500ms with a mempool that has
131072 mbufs to 10ms(regardless of the specifications).
Suggested-by: Ferruh Yigit <ferruh.yigit@intel.com>
Signed-off-by: Yangchao Zhou <zhouyates@gmail.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
When cross-compiling, if no kernel_dir was specified, then the kernel
modules were still being compiled for the build machine. Fix this by
only building modules on cross-compile when we have a kernel_dir value
set. Print out a message indicating why we are skipping kernel
compilation, and in case that the headers for kernel compile are not
found, print a warning instead of erroring out.
Fixes: a52f4574f7 ("igb_uio: build with meson")
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Luca Boccassi <bluca@debian.org>
With the legacy build system MODULE_CFLAGS can be set to pass compiler
flags specific for the kernel modules builds.
This is used currently by Ubuntu and Debian.
Set ccflags-y in the Kbuild to achieve the same result with Meson, and
to keep backward compatbility with older scripts.
Fixes regression in Ubuntu/Debian when the Kbuild is included in the
DKMS source package, as DKMS will pick it up silently by default if
present, causing the MODULE_CFLAGS to be ignored.
Fixes: a52f4574f7 ("igb_uio: build with meson")
Cc: stable@dpdk.org
Signed-off-by: Luca Boccassi <bluca@debian.org>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
The meson.build files for building the kernel modules directory could
be improved now that it is extracted from the EAL. For example, no
global processing is necessary inside the kernel folder, just need to
subdir to the appropriate bsd or linux folder to do the actual work.
To avoid potential race conditions with the BSD module builds when
the kernel build system is creating the dev_if.h and other files,
we serialize the kernel module builds (all 2 of them!) by setting
up each module to depend on all the previous.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Tested-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Starting from Linux v3.16 pci_is_bridge() is in linux/pci.h,
in previous versions it is in drivers/pci/pci.h which is private header.
Fix build error when calling pci_is_bridge by not calling/supporting
pci_is_bridge with kernel versions before 3.16.
Fixes: 6f0841b770 ("igb_uio: bind error if PCIe bridge")
Reported-by: Ferruh Yigit <ferruh.yigit@intel.com>
Signed-off-by: Scott Branden <scott.branden@broadcom.com>
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
This patch moves the kernel modules code from EAL to a common place.
- Separate the kernel module code from user space code.
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Tested-by: Bruce Richardson <bruce.richardson@intel.com>
