Ivan Boule [Fri, 29 May 2015 08:56:25 +0000 (10:56 +0200)]
ethdev: add multicast address filtering
With the current PMD API, the receipt of multicast packets on a given
port can only be enabled by invoking the "rte_eth_allmulticast_enable"
function.
This method may not work on Virtual Functions in SR-IOV architectures
when the host PF driver does not allow such operation on VFs.
In such cases, joined multicast addresses must be individually added
in the set of multicast addresses that are filtered by the [VF] port.
For this purpose, a new function "set_mc_addr_list" is introduced
into the set of functions that are exported by a Poll Mode Driver.
Signed-off-by: Ivan Boule <ivan.boule@6wind.com> Acked-by: Stephen Hemminger <stephen@networkplumber.org>
[Thomas: export new function in .map] Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Negotiate the virtio ring size. The host may allow for very large
rings but application may only want a smaller ring.
Conversely, if the number of descriptors requested exceeds the virtio
host queue size, then just silently use the smaller host size.
This fixes issues with virtio in non-QEMU envirionments.
For example Google Compute Engine allows up to 16K elements
in ring.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Changchun Ouyang <changchun.ouyang@intel.com>
Change the features from bit mask to bit number. This allows the
DPDK driver to use the definitions from Linux (yes the header
files already use a license compatiable with DPDK). This makes DPDK
driver handle future feature bit changes.
Get rid of double negative code in the feature bit intialization.
Instead just have a new define with the list of feature bits implemented.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Changchun Ouyang <changchun.ouyang@intel.com>
Don't attempt to set the MAC address table unless the host allows
it in feature negotiation. Also, don't return a value from mac_table_set
since all callers ignore the return value.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Changchun Ouyang <changchun.ouyang@intel.com>
Upcoming drivers will need to be able to support other bus types.
This is a transparent change to how struct eth_driver is initialized.
It has not function or ABI layout impact, but makes adding a later
bus type (Xen, Hyper-V, ...) much easier.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Yong Wang <yongwang@vmware.com>
The combined lib was being created after building the lib root dir.
With the new directory hierarchy, it should be created after the
drivers root dir instead.
Fixes: 980ed498eb1dd0 ("drivers: create new directory") Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com> Tested-by: Li Wei <lw@cn.fujitsu.com>
Helin Zhang [Thu, 4 Jun 2015 06:54:32 +0000 (14:54 +0800)]
i40evf: fix jumbo frame support
It wouldn't check the configured maximum packet length, and then
the scattered receiving function wouldn't be selected at all even
if it wants to receive a jumbo frame. The fix is to select the
correct RX function according to the configurations.
Add coding standards document to guides directory. This document
codifies the current DPDK C coding conventions, to make it easier for
contributors to see the format their code should be in.
Signed-off-by: Siobhan Butler <siobhan.a.butler@intel.com> Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
I run the test by:
sudo ./testacl -n 2 -c 4 -- --rulesf=./test_data/rule1
--tracef=./test_data/trace1
The result shows that the packet matches the second rule, which is wrong.
The dest port of the pkt is 53, so it should match the third rule.
"
Indeed there is problem at ACL build stage.
Sometimes acl_merge_trie() is too aggressive in trying to conserve
space at build time.
So it takes a wrong assumptions and didn't duplicate a node,
even when it should.
The easiest and safest fix seems to always duplicate a left non-root/non-leaf
node first, and let the further code to destroy the node, if it is not needed.
Reported-by: Zi Hu <huzilucky@gmail.com> Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Bruce Richardson [Wed, 22 Apr 2015 13:31:55 +0000 (14:31 +0100)]
eal/linux: fix numa node detection
Using the "physical_package_id" as a fallback for determining the
numa node of a core tends to be unreliable. Fix this by using a
detection routine which reads the numa information from
/sys/devices/system/node and just returns a numa node of 0 on
failure.
Reported-by: Wang Sheng-Hui <shhuiw@gmail.com> Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Bruce Richardson [Fri, 29 May 2015 14:34:14 +0000 (15:34 +0100)]
ip_frag: fix build with gcc 5.1
On Fedora 22, with GCC 5.1, errors are reported due to array accesses
being potentially out of bounds. This commit fixes this by adding in an
extra bounds check to the loop counter.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com>
Bruce Richardson [Fri, 29 May 2015 14:34:16 +0000 (15:34 +0100)]
null: fix build with gcc 5.1
On Fedora 22, with GCC 5.1, errors are reported due to array accesses
being potentially out of bounds. This commit fixes this by adding in an
extra bounds check to the loop counters, or, in the case of stats reset,
by blindly zeroing the whole array, rather than just the part that is in
use.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com>
Bruce Richardson [Fri, 29 May 2015 14:34:15 +0000 (15:34 +0100)]
i40e: fix build with gcc 5.1
On Fedora 22, with GCC 5.1, errors are reported due to array accesses
being potentially out of bounds. This commit fixes this by adding in an
extra bounds check to the loop counter.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com>
Bruce Richardson [Fri, 29 May 2015 14:34:13 +0000 (15:34 +0100)]
mem: fix build with gcc 5.1
On Fedora 22, with GCC 5.1, errors are reported due to array accesses
being potentially out of bounds. This commit fixes this by ensuring the
bounds check in the loop takes account of the array size.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com>
Jijiang Liu [Thu, 28 May 2015 01:36:18 +0000 (09:36 +0800)]
config: increase max queues per port
The default value of RTE_MAX_QUEUES_PER_PORT is 256, which is too small
for some configurations for i40e. There will return an error when
configured queue number is larger than 256 in rte_eth_dev_configure().
For example, in vHost sample, PF queue number: 64,
configured vmdq pool number: 63, each vmdq pool has 4 queues,
there will be required 316 queues in a port.
Signed-off-by: Jijiang Liu <jijiang.liu@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
Bruce Richardson [Mon, 25 May 2015 12:23:54 +0000 (13:23 +0100)]
kni: fix missing header dependencies
The file rte_kni.h depends upon a number of other headers, some of which
are missing from the #include lines. The following #includes are added:
* rte_memory.h - for the definition of phys_addr_t
* rte_mempool.h - for the definition of mempool struct and the mempool
create function.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Marc Sune <marc.sune@bisdn.de>
Bruce Richardson [Mon, 25 May 2015 12:23:53 +0000 (13:23 +0100)]
eal: fix missing header dependency
rte_pci.h depends upon stdio.h for the definition of the FILE type. Add
in #include <stdio.h> to the file to satisfy this dependency in cases
where the including C file does not already include stdio.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Marc Sune <marc.sune@bisdn.de>
mempool: fix pages computation to determine number of objects
In rte_mempool_obj_iter(), when element boundary coincides with page boundary,
even if a single page is required per object, a loop checks that the next page
is contiguous and drops the first one otherwise.
This commit checks subsequent pages only when several are required per object.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Reviewed-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Adrien Mazarguil [Mon, 25 May 2015 16:27:45 +0000 (18:27 +0200)]
mempool: fix returned value after counting objects
rte_mempool_xmem_usage()'s return type is ssize_t which has the same
architecture-dependent width as size_t but is signed.
On 64-bit architectures, returning a negative uint32_t value without casting
to ssize_t first does not work as intended, the sign bit is lost and the
returned value is garbage.
This commit fixes an assertion failure in testpmd on 64 bit architectures
when combining --no-huge and --mp-anon outside of Xen Dom0:
PANIC in mempool_anon_create():
line 170 assert "elt_num == mp->size" failed
Erik Ziegenbalg [Wed, 20 May 2015 00:04:55 +0000 (17:04 -0700)]
cmdline: fix small memory leak
A function in cmdline.c has a return that does not free buf properly.
Signed-off-by: Erik Ziegenbalg <eziegenb@brocade.com> Acked-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: John McNamara <john.mcnamara@intel.com>
John McNamara [Mon, 18 May 2015 11:34:07 +0000 (12:34 +0100)]
doc: refactor table numbers into references
This change adds automatic table references to the docs. The
table numbers in the generated Html and PDF docs are now
automatically numbered based on section.
Requires Sphinx >= 1.3.1.
This change:
* Adds a RST table:: directive to each table caption.
* Indents the tables to the required directive level.
Signed-off-by: John McNamara <john.mcnamara@intel.com>
John McNamara [Mon, 18 May 2015 11:34:06 +0000 (12:34 +0100)]
doc: refactor figure numbers into references
This change adds automatic figure references to the docs. The
figure numbers in the generated Html and PDF docs are now
automatically numbered based on section.
Requires Sphinx >= 1.3.1.
The patch makes the following changes.
* Changes image:: tag to figure:: and moves image caption
to the figure.
* Adds captions to figures that didn't previously have any.
* Un-templates the |image-name| substitution definitions
into explicit figure:: tags. They weren't used more
than once anyway and Sphinx doesn't support them
for figure.
* Adds a target to each image that didn't previously
have one so that they can be cross-referenced.
* Renamed existing image target to match the image
name for consistency.
* Replaces the Figures lists with automatic :numref:
:ref: entries to generate automatic numbering
and captions.
* Replaces "Figure" references with automatic :numref:
references.
Signed-off-by: John McNamara <john.mcnamara@intel.com>
John McNamara [Mon, 18 May 2015 11:34:08 +0000 (12:34 +0100)]
doc: add sphinx numref compatibility fallback
This change adds some simple handling for the :numref: directive
for Sphinx versions prior to 1.3.1. This allows the Guides
documentation to be built with older versions of Sphinx and still
produce reasonable results.
The patch replaces the :numref: reference with a link to the
target (for all Sphinx doc types). It doesn't try to label
figures/tables.
Full numref support with automatic figure/table numbering and
links can be obtained by upgrading to Sphinx 1.3.1 or later.
Signed-off-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Bruce Richardson [Fri, 15 May 2015 15:57:00 +0000 (16:57 +0100)]
xenvirt: move drivers/net/
Move xenvirt PMD to drivers/net directory
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Bruce Richardson [Fri, 15 May 2015 15:56:59 +0000 (16:56 +0100)]
vmxnet3: move to drivers/net/
Move vmxnet3 PMD to drivers/net directory.
As part of the move, rename the "vmxnet3" subdirectory, containing the
original FreeBSD drivers, from "vmxnet3" to the more standard name
"base", to indicate it contains the base drivers used for the
implementation.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Bruce Richardson [Fri, 15 May 2015 15:56:58 +0000 (16:56 +0100)]
virtio: move to drivers/net/
Move virtio PMD to drivers/net directory
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Bruce Richardson [Fri, 15 May 2015 15:56:57 +0000 (16:56 +0100)]
ring: move to drivers/net/
Move ring PMD to drivers directory
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Bruce Richardson [Fri, 15 May 2015 15:56:56 +0000 (16:56 +0100)]
pcap: move drivers/net/
Move pcap pmd to drivers/net directory
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Bruce Richardson [Fri, 15 May 2015 15:56:55 +0000 (16:56 +0100)]
null: move to drivers/net/
Move null PMD to drivers/net directory
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Bruce Richardson [Fri, 15 May 2015 15:56:54 +0000 (16:56 +0100)]
mlx4: move to drivers/net/
move mlx4 PMD to drivers/net directory
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Bruce Richardson [Fri, 15 May 2015 15:56:53 +0000 (16:56 +0100)]
ixgbe: move to drivers/net/
move ixgbe PMD to drivers/net directory.
As part of the move, we rename the ixgbe directory, containing the
ixgbe "base driver" code, from "ixgbe" to "base".
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Bruce Richardson [Mon, 18 May 2015 11:05:12 +0000 (12:05 +0100)]
i40e: move to drivers/net/
Move i40e PMD to drivers/net directory.
As part of the move, rename the "i40e" directory, containing the "base
driver" code, from "i40e" to "base".
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Bruce Richardson [Fri, 15 May 2015 15:56:51 +0000 (16:56 +0100)]
fm10k: move to drivers/net/
move fm10k PMD to drivers/net directory
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Bruce Richardson [Fri, 15 May 2015 15:56:50 +0000 (16:56 +0100)]
enic: move to drivers/net/
move enic PMD to drivers/net directory
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com>
[Thomas: move vnic/ to base/] Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Bruce Richardson [Fri, 15 May 2015 15:56:49 +0000 (16:56 +0100)]
e1000: move to drivers/net/
Move e1000 pmd to drivers/net directory
As part of move, rename "e1000" subdirectory, which contains the code
from the "base driver", to "base".
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Bruce Richardson [Fri, 15 May 2015 15:56:48 +0000 (16:56 +0100)]
bond: move to drivers/net/
Move bonded ethdev pmd to drivers/net
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Bruce Richardson [Fri, 15 May 2015 15:56:47 +0000 (16:56 +0100)]
af_packet: move to drivers/net/
move af_packet pmd to drivers/net directory
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Bruce Richardson [Fri, 15 May 2015 15:56:46 +0000 (16:56 +0100)]
drivers: create new directory
Add a new top-level "drivers" directory to which all PMDs will be moved
for easier maintenance of both lib folder and drivers themselves. This
new directory is a dependency of all the apps in the app folder, so
the makefiles for each app are updated.
To the new top-level directory add a "net" subdirectory to classify
more specifically our existing PMDs as ethernet drivers
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com>
[Thomas: fix dependencies and merge several patches] Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Bruce Richardson [Fri, 15 May 2015 15:56:45 +0000 (16:56 +0100)]
mk: use relative rather than absolute pmd paths
In the Makefiles for the PMDs, the paths to the files are often
specified using the full path from $(RTE_SDK) variable. These paths can
be shortened, and make more flexible in case of a future path change by
specifying the paths using $(SRCDIR) instead.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com>
Olivier Matz [Mon, 18 May 2015 08:18:00 +0000 (10:18 +0200)]
lib: fix 32-bit build with debug enabled
When debug is enabled for 32 bits targets, it triggers some format
errors that are not visible in 64 bits. Fix them by using the proper
format from inttypes.h or the proper cast.
Olivier Matz [Mon, 18 May 2015 08:18:01 +0000 (10:18 +0200)]
examples: add dependencies for timer and vm_power_manager
Do not compile these examples if the related dpdk option is not
enabled, as it's done for other examples. It allows to build
the examples directory with a reduced dpdk configuration.
Olivier Matz [Mon, 18 May 2015 08:17:59 +0000 (10:17 +0200)]
examples/netmap: fix build for x32 ABI
Fix a cast issue:
examples/netmap_compat/lib/compat_netmap.c:827:10: error: cast to
pointer from integer of different size [-Werror=int-to-pointer-cast]
Thomas Monjalon [Tue, 19 May 2015 09:34:16 +0000 (11:34 +0200)]
examples/ip_pipeline: fix build with clang 3.6
This error is detected:
examples/ip_pipeline/cmdline.c:272:15: error: address of array
'params->file_path' will always evaluate to 'true'
if (!params->file_path) {
~~~~~~~~~^~~~~~~~~
file_path is an array in a structure so it's unneeded to check it.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com> Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Helin Zhang [Mon, 18 May 2015 15:40:56 +0000 (23:40 +0800)]
i40e/base: fix build with clang
When compiling with clang, it detects that a variable used for debug
is unused (debug_mask is never set):
lib/librte_pmd_i40e/i40e/i40e_nvm.c:708:20: error: unused variable
'i40e_nvm_update_state_str' [-Werror,-Wunused-variable]
STATIC const char *i40e_nvm_update_state_str[] = {
^
As we are not supposed to modify the base driver (except for update
synchronization from Intel), it was ignored with GCC thanks to the
option -Wno-unused-variable but clang was forgotten.
Miscellaneous modifications are added as follows for future use.
- Admin queue command opcodes
- struct i40e_aqc_arp_proxy_data
- struct i40e_aq_set_phy_config
- struct i40e_aqc_nvm_config_data_immediate_field
- struct i40e_aqc_oem_param_change
- struct i40e_aqc_opc_oem_ocsd_initialize
- struct i40e_aqc_opc_oem_ocbb_initialize
In addition, sturcture and command length check at compile time
are added for each admin queue opcodes in order to report any
potential issues as early as possible.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
Several structures as follows were modified for future use.
- struct i40e_link_status
- struct i40e_nvm_info
- struct i40e_veb_tc_stats
In addition, shadow RAM macros and enum were modified for future
use as well.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
Make sure the sizeof() calls are taking the size of the actual
struct that we care about. By using the pointer variable, it will
always get the right struct size, even if the variable type
changes sometime in the future.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
Four new interfaces are added for future use, as follows, together
with their relevant strucure and macro definitions.
- i40e_aq_read_nvm_config()
- i40e_aq_write_nvm_config()
- i40e_aq_set_lldp_mib()
- i40e_read_pba_string()
In addition, removed i40e_read_nvm_srrd() as needed.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
i40e/base: workaround for PHY type with firmware < 4.4
The FW for SW3E is incorrectly reporting a breakout cable as PHY
type 0x3 when it should be 0x16 (I40E_PHY_TYPE_10GBASE_SFPP_CU).
As this is supposed to be fixed in the next FW version
(FW versions >= 4.4), if we get this value back from FW and the
version is < 4.4, reassign it to I40E_PHY_TYPE_10GBASE_SFPP_CU.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
i40e/base: find partition id in NPAR mode and disable FCoE
In NPAR mode the driver instance might be controlling the base
partition or one of the other "fake" PFs. There are some things
that can only be done by the base partition, aka partition_id 1.
This code does a bit of work to find how many partitions are there
in per port and what is the current partition_id. In addition,
FCOE is disabled by default, as it was pushed out from the firmware.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
Getting the pf_id from the function number was a good place to
start, but when the PF was setup in pass-thru mode, the PCI
bus/device/function was virtualized and the number in the VM is
different from the number in the bare metal. This caused HW
configuration issues when the wrong pf_id was used to set up the
HMC and other structures. The PF_FUNC_RID register has the real
bus/device/function information as configured by the BIOS, so use
that for a better number.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
As i40e_update_link_info() is just a simple wrapper of AQ command
interface of i40e_aq_get_link_info(), the AQ command interface
should be used directly.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
The timeout of ASQ command is enlarged from 100 ms to 250 ms to
wait longer time for the response. nvm_busy is not needed anymore.
Also nvm_release_on_done and nvmupd_state will be set during
adminq init.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
i40e/base: use FLAG_RD when sending driver version to firmware
When sending driver version to FW, FLAG_SI is not needed as it is
done as part of i40e_fill_default_descriptor function call, and it
should be replaced by FLAG_BUF and FLAG_RD.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
i40e/base: grab NVM devstarter version instead of image version
The NVM image version has useful data but it is a per image version,
and every image can have a different one. While for the dev starter
version, all the images for release will have the same version. So
it would be better to use dev starter version number.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
The data buffer for i40e_calc_nvm_checksum() is too big, so move
it off and allocate separately. In addition, use
i40e_read_nvm_buffer() to get the whole shadow RAM, together with
minor enhancements.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
The state transitions after an error were not managed well, so
these changes get us back to the INIT state or don't transit
out of the INIT state after most errors.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
To support better debug information printing, all DEBUGOUT() are
replaced by i40e_debug(). In addition, the NVM update state strings
are added in debug information.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
Decoding the AQ return code is great except when the AQ send timed
out and there's no return code set. This changes the handy decoder
interface to help catch and properly report the condition as an
useful error number rather than returning a misleading '0'.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
i40e/base: catch NVM write semaphore timeout and retry
In some circumstances, a multi-write transaction takes longer than
the default 3 minutes timeout on the write semaphore. If the write
failed with an EBUSY status, this is likely the problem. So here
it tries to reacquire the semaphore and then retry the write.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
To support different hardware, shadow RAM read should be done via
SRCTL register or AQ command, and shadow RAM write should be done
via AQ command only.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
i40e/base: rework virtual address retrieval for lan queue
To simplify the usage of invoking 'i40e_hmc_get_object_va', input
parameters of it were reworked. In addition, minor code style fixes
were added as well.
Test report: http://www.dpdk.org/ml/archives/dev/2015-May/017384.html
Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jijiang Liu <jijiang.liu@intel.com> Tested-by: Min Cao <min.cao@intel.com>
git.droids-corp.org / dpdk.git / log